I'd rather have flagship specs in a smaller package. The smallest in the iPhone (SE) and Pixel ("A"?) lines are still too big and tend to have previous-gen specs
Because the execution is usually borked... I was eyeig ZenFone 9 (or something around that) and what? It was reported that it had problem with overheating and build quality.
What's more, I would love something akin to my current Galaxy a52s 5G with a display around 5.2-5.5" (I first had LG G2, then OnePlus3 which was already a bit bulky and now a52 as compromise; https://www.gsmarena.com/compare.php3?idPhone1=5543&idPhone2...)...
I do have iPhone SE (2022) and it's the size of LG G2 and I find quite handy. Something of that size but with slightly bigger screen (better screen-to-body ratio). Specs doesn't have super-hiper-premium… and the price should be sane (usually compact phones are like 20-40% higher, sic)
In my case, they took so long to announce the iPhone 12 Mini that I gave up waiting and bought an SE even though it was slower than I wanted and had a poor camera. Four months later they announced the Mini, but I wasn't willing to replace a four-month-old phone. Then they discontinued the Mini line after 13.
When I was ready to buy a new phone, there were no iPhone Mini models for sale. It took more than a year, but I finally found an iPhone 13 Mini in stock on the Apple Refurbished store. Now I'm hoping to keep this phone alive until they finally release another small iPhone.
Are you sure they bought different flavours because of the screen size or because of naming marketing "pro is better than regular" and "mini obviously has to be worse"?
Did you know that factories make less big-sized and small-sized shoes than average-sized ones? Because (surprise) buyers size distribution is not uniform.
“We tried to make big shoes many times and it doesn’t sell well enough”. Oh, really. I guess I’ll just cut holes for my fingers then.
Maybe the fixed costs of a shoe factory production line, in 2024, with centuries of production experience, are lower than those of a top of the line smartphone.
Speaking for myself, I 100% would have bought an iPhone Mini, but I purchase new phones on a 5-6 year cycle. My iPhones were the 4S and the XS.
Now that I'm ready to buy a new iPhone, the Mini has been discontinued! I think the Mini would have been and in fact was successful, but it's not successful "enough" to justify a separate model – they must have observed that people "like me" would still buy a flagship iPhone, even though we aren't 100% satisfied with the form factor.
Apple would rather have us buying a higher-margin flagship model and have an NPS of 65+ than a lower-margin mini model with an NPS of 80+.
My friend with similar instincts as me recently got a refurbished 13 Mini instead of the latest flagship. I'll probably get the flagship, because I value the satcom a little bit more than the form factor.
That’s one of the big problems with these dominant trillion dollar corporations. For them it’s not worth to pursue something as small as a billion dollar market but they still have enough market power to suppress any competitor who wants to deliver such a product. That’s why we are seeing less and less innovation and diversity compared to the 1980s to around 2000
It’s a package of devices Apple is selling. They sell under powered smaller screen macs. They sell iMacs. They sell small iPads. They sell smaller watches. Some of these just don’t sell as well as their other offerings.
They should sell small phones. Because the whole family will be in on the brand, features, and services. I’ve heard many family members and friends say that they won’t give up their older small phones because Apple no longer makes new ones.
The idea that they don’t sell well is not a good enough reason when you are trying to capture the whole market for not just hardware but the lock in for services, apps, games, music, etc.
How can you be confident in this when you don't even see their sales data?
Still, I would like to see a smaller regularly updated phone. Bonus points if there is a high-end version, because small shouldn't mean budget (like with iPhone SE).
The last small flagship (iPhone 13 mini) sold poorly, but it was much more expensive than the SE2. This was at the tail end of covid, but before faceid worked with masks, so the SE's touch id was a huge selling point.
Other than that and the camera, the only functional difference I can find are that the SE line is still missing the UWB antenna.
I'd happily upgrade to a newer small iPhone if they made one. As it is, it looks like the only option is repeatedly repairing my 13 mini (and dealing with the hilariously bad 5G battery drain forever) or downgrading to a newer SE3.
I know there are a lot of people in this boat. I predict they'll produce another small phone in a few years. It'll sell well due to pent up demand, and someone will be declared a genius for selling 100M's of extra phones that year.
> I predict they'll produce another small phone in a few years. It'll sell well due to pent up demand, and someone will be declared a genius for selling 100M's of extra phones that year.
I’d wish this too. I’m afraid that Apple over the next few year would become more risk averse then ever before.
Also old Execs are leaving and retiring - so less people with hands on experience how to start new products VS keeping lights on.
"Well enough" is doing a lot of work here. It's not that they aren't successful, it's that they aren't immediate, runaway hits, so the manufacturers conclude: why bother trying to build this market, let's just go back to the playbook. That's how you get mediocre products, which is where we are now.
I'm sorry to say this, but again just face the reality of the fact that not enough people truly want this size of phone.
There's a congregation of people on HN who do, I suspect they're also the type of people who'll run their phones for 5+ years where their bigger phone buying brethren are replacing every year or two at most.
The market for the "small" smartphone just isn't profitable enough to bother with for most manufacturers.
> I’ve got an appointment with Apple to replace the battery in my iPhone 13 mini.
I literally was in the Apple store yesterday for the same purpose (with a 12 mini). I'd also love the new features and hardware, but after trying all the available sizes in the store, they're all too big.
My wife on the other hand(s), loves to have a phone she needs to hold with two hands to even be able to use, so obviously she has the Pro Max. I don't understand how people are OK with that, but to each and their own...
Good luck. I’ve tried to replace battery in mine 12 mini last week - with no success. I had to leave my phone for 4 hours or several days (if they brake screen during battery replacement, they will wait for replacement phone to be shipped overnight). Also representative was convincing me to buy a new phone - saying that battery replacement won’t help much because new ios versions has features which high battery usages, while newer iphones has larger battery and hardware optimizations for these new features. I’m thinking about iPhone 16 now while keeping iPhone 12 mini as backup phone.
I've successfully have replaced batteries and displays in older iphones (mainly iphone 6). But with newer iphones opening the phone is more complex.
I've read online and heard from Apple Store representative that iPhone 12 (all models) has tendency to crack the screen when phone is opened for repair or battery replacement and in that case Apple Store would replace the hole phone (this is were multi day repair process).
So I would rather pay $90 to Apple that guarantees that I'll get a phone replacement in case when screen is broken during battery replacement. Without the phone I sill would be able to answer the cell calls from Apple Watch and with ipad over WiFi.
They would tell you and you would notice if your phone was replaced - it would have new serial and 2FA apps won't work on a new phone without reregistration.
Most likely it took them a week to get a new battery for replacement shipped.
My opinion is that most of the real uses of AI (like ML has always been) will be largely hidden things that are LLM based but not screaming at your face "AI". Particularly once the bubble pops and money stops being shoved into things just sticking an LLM in a pretty package with little to no value.
Some of the things coming in iOS like notification summaries and similar features are big examples. It's clearly LLM based but it's not a lot of the shoving AI needlessly into things that we are seeing now and provides a true improvement given the notification overload that we have right now.
Exactly. Ask customers "do you want AI in your phone?" and their response will probably be "meh", as shown in the article. But ask "do you want notification summaries, a better camera in low light, Siri to be able to look up more things, searchable photos, etc?" - and they absolutely will.
You can see the wisdom of how Apple is approaching this. In particular, to be on device whenever possible so as not to be dependent on network bandwidth, and to tie features to new hardware (to drive sales).
I have a hard time seeing how this isn't obvious. 95% of everyday AI needs (for the people that even bother to interact with it) are covered by ChatGPT, and most of that is the same stuff that Google was handling before.
From personal experience, the only thing that changed when replacing the "old" Google Assistant with the Gemini-powered one on my Pixel was that it's no longer able to create reminders.
The only thing that changed for me was a that it couldn't control my smart Home devices anymore, nor activate navigation, nor send messages via Whatsapp (I.e. while driving).
Literally every thing I used it for got answered via "I cannot do that yet" after it randomly opted me into that. Pure garbage.
Then why are you using it? I tried using Gemini once on my Pixel 6. Couldn't play music on Youtube music on verbal instructions, I switched back to Google Assistant. Will try it again after 6 months now. :)
I'm not, I pretty much just accepted that Google doesn't care about usability whatsoever and haven't prompted it in a very long time.
To be clear, the only time I've ever used it was via "ok Google" in contexts in which I'm unable to interface with the phone directly, i.e while driving. If it doesn't work you'll learn that you can't start driving before queueing the navigation anymore. The voice assistant was a nice feature, but not important enough to waste my time trying to figure out which feature they opted me into and how to get back out of it.
In my case it just kinda.. switched over at some point, and frankly I didn't care enough to figure out how I might switch it back (if I even could). I had a similar frustration to GP that it stopped working for 100% of the queries I used to use it for.
That said, at some point it started working better, but there was a good 6-12 months where it was a tire fire.
I'm not sure that's true. Not everybody is hounding for information from the web in the first place.
Apple's approach of using current-boom AI to help you navigate and digest your own private trove of multimedia content (photos, videos, apps, notes, structured data, etc) is absolutely useful to people as well, and for some of us, one of the only personal uses of this AI that seems compelling at all.
I'm much more excited to have help finding that goofy picture of my cat by describing what I remember, so I can share it with a friend, than I am to have some chatbot dialog about entry-level Python with a hallucinating parlor trick.
But these features have to work, and work well, and work fast, and be widely known to work, before they'll really win the market. But that's going to take a minute and it might not even happen.
>I'm much more excited to have help finding that goofy picture of my cat by describing what I remember, so I can share it with a friend, than I am to have some chatbot dialog about entry-level Python with a hallucinating parlor trick.
Hasn't Google been doing this forever? I can search random things in my photos (like pictures of an old car I owned).
For a bit, and to a degree, yes. Last-decade image recognition and tagging teased what might be possible and is genuinely useful.
The new LLM-ish tools promise that users can be more vague and casual in what language they use and more elaborate in how specific they mean to be; and that the queries (and operations) can span more diverse data sources.
iOS has been doing this for a bit too. I don't use it enough to really know how good it is but I can definitely look for cats or people I know. Haven't used Apple Intelligence yet so maybe that's better as well?
Google photos is way, way, way better than apple photos at this. It’s not even a competition.
I have my sister’s dogs named in my google photos library. Every time I a take a picture of either dog, they are automatically tagged and added to a shared album I set up for my sister.
I have nieces and nephews with photos from newborn age to 10+ years old, and it has managed to organize them across their growth and ages. It’s incredible. I can search for “<niece name> <vacation area>” and get every photo of her on a certain vacation to make a family scrap book.
Apple photos search and tagging is pitiful in comparison.
Yeah, it seems based on the advertising from the various AI vendors, they are showing its use by summarizing emails/phone call/etc. Things like being able to search text messages for info blah blah. The only one I've seen pushing online searches is Google, but that seems like duh! for them to be pushing. Circle something in an image and take me to a listing of that something for sale. Of course that's Google's direction.
But that whole find me something on my mutliple gigabytes of storage on my device account definitely seems like the mass appeal
Mundane or not, it’s actually useful, a meaningful improvement, and should work consistently well.
That is in contrast to a lot of fancy AI demos which are a great party trick, but fall apart in actual usage, with their reliability being “maybe it will work this time, maybe it won’t, just keep retrying :)”.
Apple is pretty well-recognized for usually being a bit late to the party, but at least delivering stuff that’s polished.
Just look at this thread of people sharing how Gemini broke all their commands and automations. The Apple Intelligence Siri on the other hand works just fine (even if new features are arriving slowly).
I'm not sure that a hallucinated image of something is better than the original image when you're doing spywork.
The difference between whether someone has 4 or 5 fingers, whether they're holding a gun versus a random object, or whether they're mixed-race or caucasian, all seem like they would be pretty important things. Likewise, car number plates, signage in the photo that help identify where it is, metadata of the image itself (often more useful than the image), are all incredibly important. All of those are things that AI is absolutely terrible at lmao.
The problem is, vast majority of smartphone usage is done for entertainment and social networking purposes (IG, TikTok, Twitter, HN, gaming, Netflix and etc.). If you’re mostly scrolling and consuming, I can’t imagine how current AI tooling can help you other than some summarization of texts. Sure, for productivity cases it might be legitimate, but that’s not what supermajority of people use a phone for.
the fun thing about the chatbots regularly making stuff up is that they almost always know when they're making stuff up. the hallucination problem isn't a problem of not knowing the facts, it's a problem of not knowing whether you want an accurate answer or a creative answer.
try asking chatGPT to only give you true and accurate answers and not make anything up.
I trust a defective AI a lot more than I trust Fox News ;-)
Now, more seriously, it'd need to put together a coherent argument and back it up with reputable sources, as just citing sources is very ineffective. The article I cited gives more details on possible approaches to that.
People eternally hoping that some new trick will finally make the other people understand how their side of the story/argument is one and only truly truthful. I guess this is old as mankind.
With your argument, the problem happens when given person goes to Fox news in the first place. Selection of the source has already been made, with its biases. Not much you can do or expect after this point.
Also, who curates the curator? Again an age old problem with no real, long term working solution in sight. No, you should not expect some statistical model to hold your hand through vast internet, while giving up any form of critical thinking, reasoning, or I guess any cerebral process altogether. Ultimate laziness. Since we know how much money there is in diet fad business, its safe to say this above will find its non-tiny desperate crowd.
> their side of the story/argument is one and only truly truthful.
There are no sides in objective reality. You might offer competing hypotheses and evidence for those, but we don’t need to do that to know that climate change is real, that it’s caused by humans, and that vaccines work.
That’d be useful but I’m pretty sure they’d get a massive backlash on Fox News and lawsuits filed alleging “being cancelled” within minutes of that shipping. It’s something we need but our current disinformation problem isn’t an accident but the result of decades of investment.
I can't imagine that working, even if that AI were more reliable than currently possible. Someone who already disbelieves credible reporting and objective sources is not going to be swayed by an AI telling them to disregard their fantastical conspiracy theories. Especially not in a near-future world where everyone is inundated with AI fakery.
That's even without considering the efforts that would be made to undermine any system that showed any effectiveness. A lot of misinformation, probably most of the stuff that gets traction, isn't random, it's serving a purpose and being pushed for that reason.
And who's going to teach the AI what's a conspiracy and what's not? Battles have been fought over what narrative certain Wikipedia pages should push; it's pretty hard to find information that can be objectively considered the truth.
Im not sure people really know how to use an AI chat bot or understand its a more powerful, quicker and personally more fun way of getting information then just a Google search
Some personal examples of how I find it more useful (love to hear yours) and fun to use....
- Wanted to go on a hike an hour away from where both my friend (lives two hours west of me) and I live. Asked GPT what are some good hikes an hour drive away from both of us to meet & hike. With Google I have to do Many searches where GPT just provides the answer right away.
- I count calories and eat out everyday. GPT knows the calories of everything i eat as I eat at chains mostly (Cava, Panera, Starbucks, Chipolte). I tell it via voice what i just ate for my 1st meal, it calculates my calorie count and later I'll tell it what im having for my 2nd meal. It can also recall my calorie count from days ago. It does all this quickly vs. Google i'd have to do oodles of searches.
Usually Im using GPT the most when driving via voice and unlike Siri, GPT understands me and i can have whole conversations with it to get things done while driving.
> Im not sure people really know how to use an AI chat bot or understand its a more powerful, quicker and personally more fun way of getting information then just a Google search
I can't speak to your personal entertainment experience, but AI chatbots are generally a slower, less accurate way of getting information than a google search. (Though Google polluting search results with a big, often inaccurate, AI result at the top narrows this a bit.)
If you use it for research where you have to do many google searches vs. just ask one question like hiking question it's much quicker asking one question vs. multiple google searches to get ur answer.
I don't really understand why it is acceptable to speak for others on this topic. It is fine if it doesn't work well for you. It is also fine if it works well for others.
All of that sounds like the boring low-hanging life fruit that gets trotted out in videos by companies like Apple and Google as being "revolutionary." It's boring. It's staged. It's the easy stuff. It's well-off 20-somethings solving non-problems for other well-off 20-somethings.
Wake me up when I can say things like…
Hey, Google, are my custom license plates ready for pick up at the tax office?
Hey, Siri, ask my doctor to refill this medicine.
Hey, Alexa, how many charging stations are broken at the gas station on 16th street?
Hey, Google, why is this plant dying?
Hey, Siri, why are there so many people in my neighborhood today?
Hey, Alexa, did anything ever get done about that story in the newspaper from a couple of years ago about the Chinese slave labor being used to grow pot on illegal farms on the Navajo reservation?
"AI" just doesn't have access to the information required to do anything interesting or useful. And because so much of its information comes from the web, which is already so polluted on certain subject (gardening, travel) as to be useless, the AI becomes useless.
I've heard horror stories and have held off so far in 'upgrading'. In the end I don't really want the fully flexible responses people are leaning into with these llm tools. All I want is to be able to give a precise instruction with my voice and have the machine reliably perform the action that it performed the last time I gave that instruction.
Since that seems to be an increasingly niche desire (at least as far as the product managers are concerned), I've been looking more and more seriously at setting up my own local voice assistant. My main barrier has been hardware—the mic arrays in the Home devices are surprisingly good and hard to beat with cheap off-the-shelf components, and you need a good mic for good STT.
Yeah, I would like Siri to actually play the album I asked for and not something completely different phonetically from what I asked. Or even when I set an alarm and not be told "I can't connect to the internet right now" while I'm using my laptop connected to the internet. Or if my internet is down to actually use the speakers that I bought as speakers.
The hardware is really well done but the software is either over or under-engineered to a stupid degree.
It also sometimes asks to unlock my phone for commands that plain old Assistant was happy to do while locked. I haven't really found it useful at all yet, free ChatGPT is just better than free Gemini for "LLM stuff" and Google Assistant is better for "smart home stuff"
I've been thinking about trying the OpenAI integration for home assistant[1], because controlling things in my home is primarily what I use my assistant shortcut for. The normal assistant works well enough but can be frustrating if you don't remember the exact phrasing it wants to activate a certain command.
I have it set up with ollama. It’s… interesting. HA commands are provided to the model as tools so it works as well as the model is able to determine when and how to use tools. From experimenting with that and my own tool use code I’ve found that models vary greatly in their ability to wield tools and none that I’ve tried are exceptional.
It’s neat that you can intermix general chatting with HA commands but you’re probably going to find that the old assist is more reliable for commands. What I do like is that you can use a template as your system prompt so you can provide the state of a number of entities and then ask for them with natural language. That works well.
I have an Alexa/Echo voice announcement system set up and have recently tied that into assist so I can do automations like if the garage opens I prompt for “what is the state of the garage?” and announce the result. Makes it feel more humane than the same plain announcements all the time.
Do try it. I've been running it ever since it got integrated into the core, mostly to control A/C units around our flat, and it's the best voice assistant experience I had to date.
I mean honestly, how is it possible Amazon, Apple, Google and Microsoft[0] all keep screwing this up for over a decade now? I literally spent 15 minutes hooking up GPT-4 to the Home Assistant integration, and I was able to semi-reliably[1] control actual devices[2] like air conditioners and smart lights, in a completely natural and ad-hoc way, by talking to my smartwatch on the go, or to a phone, whatever was more convenient at the moment.
It's a really magical experience, a step closer to Star Trek reality. And what makes it possible is not just LLMs being able to deal with natural language, but more importantly, "bring your own API key" model allowing to cut away all the bullshit that FAANG assistants are stuck in.
--
[0] - Ever since they dropped MS Speech API in Windows, and did the Cortana thing. Some 15 years passed, at this point, and I'd still prefer to work with the Speech API than to touch any of the FAANGs' voice assistant - it worked, and worked off-line!
[1] - Works ~90% of the time; some 5% of the time the voice model (from Home Assistant Cloud) misunderstands me, and 5% of the time the LLM gets confused. It's still worth it, because I can actually talk to it like to a person, without thinking of style or grammar or magic keywords.
[2] - Which, given the level of integration of Home Assistant companion app with the phone, can be easily turned into an equivalent of on-phone voice assistant that can do more than the one I got from Google. Critically, there are ways to couple Home Assistant app and Tasker, so it's not hard to make it do arbitrary things on your phone. And, if you don't like low-ish code Tasker experience, you can trivially shell out from Tasker to Termux, at which point sky is the limit. Point being, an enthusiastic non-developer with minimal tech aptitude can beat Google and Apple at the voice assistant game today.
AI’s appeal depends a lot on the features. Battery is important for me (more than being anorexically thin), but I would love an AI that could screen my calls like a smart voice mail that asks questions.
Also, being able to talk to my mailbox asking questions about subjects mentioned in my e-mails would be a huge time saver.
Imagine a purely local Microsoft Recall-like thing that could answer questions about things you saw, or that read the news articles you went over quickly and answer complicated questions about them much later, at a time you just started to regret not having bookmarked it for future reading.
Computational photography uses AI a lot (next to parametric approaches), and without these algorithms, smartphone cameras would be quite shitty. People do care about that. AI isn't just LLM chatbots.
I know that's true, but I find that the images on my Pixel are starting to have a bit of an eery feel, with some of the details looking more and more like AI generated images. I'd give back a bit of the quality for more "natural" looking images.
I'm having a horrifying realization that all of my pictures are "fake" in the sense that they don't match what I saw/experienced. Maybe it's time to get back into Polaroids.
> I'm having a horrifying realization that all of my pictures are "fake" in the sense that they don't match what I saw/experienced.
I'd caution that judicious/proper post-processing is actually needed if you want that result, because of the differences between the sensors.
Your human experience comes from many small pictures taken by a set of lenses panning across multiple points in a scene with constantly adjusting exposure times and focal lengths, all biologically composited into what feels like a single moment.
Trying to fully replicate that with a single artificial picture is going to be deficient in certain ways.
---
Separately, a pet peeve of mine: Too many people have been subtly brainwashed into conflating the "like I was really there" with " like a Hollywood film camera was really there." Then the next thing you know your medieval fantasy game has lens flares in it for no good reason.
I recently bought a real camera partially for this reason. I don't even mind how inconvenient it can be at times because honestly taking photos with a full camera rather than your phone is fun. While on the phone it has become quite dull in my opinion.
But knowing that the phone does a lot of software tweaking to get a picture to look similar to how good a full camera is made me want to switch. I think this was around the time that article about Samsung basically replacing a photo of the moon.
What does this mean? Sure, if generative AI is filling in features that weren't present, anyone would call that doctored or fake. But computational photography is mostly about recognizing patterns and filling in assumed detail... which is also how your visual cortex works.
Get raw, and marvel at its imperfection and ugliness in many aspects. If given phone ain't giving up raw data, take any decent camera, raw sensor data has been part of it since its beginning.
But those pics will be probably further from perceived reality than those enhanced by software (lets not get retarded here and don't brand every data processing as 'AI'). Distortions not only of barrel type, waning brightness towards edges, moire, heavy vignetting, tons of noise, over/underexposition, maybe some dead pixels... thats not how I see my days go by.
What makes a Polaroid any more "real" than an iPhone picture to you? Can any photo truly be real? (Deleuze has some interesting thoughts on the matter)
Interpretation of this on its nose suggest the algorithms are the core feature not AI, as there is no artificial intelligence involved in these processes that I’m aware of.
If you actually peak under the hood they just pass through weighted selectors, no different than a switch statement
what about the infamous recognizing the moon in the background that is just an over exposed white fuzzy circle and replacing with a stock image with full surface details? clearly there's some sort of ML/recognition of content within the image
Yes that would be ML as far as I’m aware, but the thing in reference is software compensation for image quality. Effectively smartphones automatically upscale photos by default. I think the only exception is when you choose to shoot in raw
This assumes that the capabilities and use cases are unchanged. Yes, for the AI features available today, I suppose ChatGPT can do much of it -- I wouldn't know because it's not interesting or useful to me, so I don't use it.
But: If I'm deciding whether AI features are important to me in making a decision to spend money on a future phone, it's those future AI features that I will be assessing.
95% of my everyday needs for an external intelligence (besides my own) are covered by e-mail, text, and phone calls with other humans, with a trivial portion covered by nascent AI features. As this changes, and AI gets more capable of replacing human intelligence in these interactions (TBD if this happens in the next smartphone generation, or the next human generation, or further in the future), then I will /very much/ care that the electronic device that I use most often day-to-day has access to these capabilities, and will very much use access to those capabilities as part of deciding where to spend my money.
You seriously assess "future AI features" when buying a phone? Have you heard the expression "a bird in the hand is worth two in the bush"? Also, what is the lifetime of a phone? How far in the future are these new features expected?
No, I assess current AI features when buying a phone, in the future. Any survey about what features customers value in phones they are considering buying inherently ask questions about future behavior, occurring in the future’s present.
Apple will just tell you you need an iPhone 17 because it has a more special "you're gonna love it" neural thingy onboard, and so your purchase of a 16 is null.
> 95% of everyday AI needs (for the people that even bother to interact with it) are covered by ChatGPT
True, but it is an external service with all the privacy concerns that entails. I appreciate eg. Apple pushing local AI but at the same time I don’t think it needs to ship with the OS. Just provide an AI API apps can hook into then I can decide which models I want and where they run.
> the only thing that changed when replacing the "old" Google Assistant with the Gemini-powered one on my Pixel was that it's no longer able to create reminders
Weird! Mine can still create reminders, as well as set timers and alarms.
AI is usually about implementation details. Customers typically worry about solutions to problems. They don't care how solution is implemented - using AI or not.
Perhaps a step further: I do not need my phone to tell me an LLM-generated bedtime story involving a specified cast of characters, I just want it to reliably control existing functionality, dangit.
Like when a timed alarm is making the phone buzz in my pocket while I'm driving, I'm telling it to silence the alarm, and the response-voice is regretfully informing me that there are no alarms going off right now.
"Everyday AI" hasn't even been built yet tbh. Where is my assistant that notices my flight has been delayed and reschedules my appointments and lets my contacts know?
I think your example highlights why this kind of "everyday AI" would be nearly impossible. If my phone automatically rescheduled my appointments and notified my contacts, without my explicit direction, I would be pissed as all hell. There are some "confirmation" notifications that could assist here ("We noticed your flight is delayed - would you like to reschedule these appointments?"), but even then, I'd say the majority of notifications I get these days are annoying and I spend a ton of time trying to figure out how to turn off annoying notifications while not totally silencing ones I depend upon. I'd have a difficult time believing that AI systems wouldn't just add to my list of over-burdensome notifications.
If you were an important CEO or something, and you had a competent human Personal Assistant - would you allow them to reschedule your appointments if you were running late?
Seems to me the problems are (1) the "assistants" aren't anywhere near good enough to be trusted to make the right decisions, and (2) a trustworthy assistant isn't compatible with the adtech business model, so it's unlikely facebook or google would produce such a thing.
>If you were an important CEO or something, and you had a competent human Personal Assistant - would you allow them to reschedule your appointments if you were running late?
Yes, why wouldn’t I? I could also give them parameters like “if I’m more than 20 minutes late please re-schedule this” or “if my flight is delayed please let everyone know it’s delayed”
Why wouldn’t I do that? Presumably the person hired is competent to make determinations within parameters specified.
I could also let them know when it’s inappropriate to do this. Again, they should be competent enough to discern the differences between when it is and isn’t appropriate.
This could honestly be done by an algorithm if you give it the correct inputs and outputs and it could be fed updates, the only real limit is the fact that some of this isn’t exposed via an API either in a timely fashion or at all
Important CEOs bring their trusted human assistant with them. Those of us at a lower level don't get that option, but at the CEO level you get to hire your own sectary and you spend years teaching them how you work and what to do in all the situations that come up and so they will make the right decisions often enough that you can trust them. Or they know they are not trusted to reschedule that appointment and so they don't.
That years of training is what we are missing. I don't think modern AIs can be trained in the way the assistants of old could be, at least not yet.
I don’t even think you need machine learning for this, it’s an API problem mostly like I said.
Being able to collate the requisite inputs from outside sources is the real problem. If you can’t do that reliably it’s simply hard to build an algorithm around it. Flights for example would require your calendar program to reliably pull data from an API regarding the flight information that is current and effectively real time. That’s the actual hard part, and this expands across services.
For all the advances we have made with computers and smartphones in particular they suck at meaningfully exposing a way to collate data sources and create actions around them reliably
Personalizing any automation like that is a privacy nightmare. The system needs to know a lot about your preferences and the decisions you'd normally make yourself as well as your current circumstances since those will also influence your choices. How do you feed that into any AI without being problematic for privacy?
Having one running locally helps but it's still necessarily storing information that you might not want to have stored where someone could potentially retrieve it, either via some sort of exploit or by forcibly compelling you to give it up.
I worked on a (failed) system that was supposed to help with this in the pre-ChatGPT era. The obvious limitations is getting the data, and scaling that to everyone. Learning about your flights, staying up to date on them, etc is such a daunting privacy-busting task that everyone is scared to start. Either you go use-case by use-case (flights, restaurants, calendars… etc) and never get traction or you just start scanning emails and open up huge data risks.
Today, this is nearly available, nearly. Probably only something Google/apple can realistically offer. Apple “intelligence” has started to read your notifications and rewrite them for you, so it shouldn’t be a big leap to listen for a United App notification and decide it’s urgent enough take action. Should be “trivial” for Google to do as well, and they could even run it server side to help without a phone present.
I love how the goal post is being moved publicly as to what AI means.
AI is anything automated it seems, and now they’re being subcategorized into niches as to what they do, e.g. “Agentic AI”, “LLM backed AI systems” etc.
If it’s not real intelligence then it isn’t really AI, and I wish the world at large would call it out.
LLM, Machine Learning, Neural Networks etc are all great but none of them have true spontaneous intelligence or learning ability.
Please, someone point out how any of these systems have organic spontaneous learning ability for a subject it was not pre-data seeded on. This is a generally accepted measure of higher level sentience as far as I’m aware
None of these exhibit any accepted definitions of intelligence markers. The intelligence isn’t artificial, it doesn’t exist is my point. If you apply the commonly accepted definitions of what is considered display of intelligent behavior. One aspect of which is the ability to adapt to new circumstances that you haven’t experienced before.
There has been considerable success in programming computers to draw inferences, for example, but not actual reasoning. You can mimic some forms of reasoning but you can’t take one ML set - like recognizing photos with mountains, then expect it to correctly identify a similar geographical element - a hill. It can’t do that. It may correctly identify that it’s not a mountain but that isn’t the same thing as actually learning it’s similar to a mountain but not the same, which would be a rudimentary definition of a hill that an intelligent entity could conceivably use if it knew what a mountain was but not a hill.
Machine Learning was always a more honest place to have This discourse. I am indeed pushing back on the idea that we should be calling ChatGPT or anything like it intelligence.
It’s Machine Learning, clever algorithms, Large language Models, among other things, that are trained on ways to mimic certain aspects of intelligence, but it does not actually possess any real intelligence. Look at the LLM hallucination problem for example. It can’t be self corrected because it’s not an intelligent system.
Moving the goal post on what AI means (and pushing AGI as some new goalpost) is disingenuous, and relatively recent.
I’d care not if it wasn’t for the fact there is so much misinformation around capabilities and the future of AI, that it’s already negatively crept into policy making for example.
1. How does it know how/when to reschedule that doctor's appointment that you need to call to reschedule? How do you know that receptionist won't hang up on your "AI agent" who tries to call them, b/c they think it's some sort of scam bot?
2. How does it know which contacts to contact? Does that acquaintance you talked to for some professional reason need to know your flight got rescheduled? What about that travel agency you talked to last night to confirm the flight?
I would say the problem there is the requirement to phone a human to change a record in their database.
If they had an app then an AI assistant should be able to tie things together. Where things seem to be going is apps provide an intent-based API wrapper plus UI widgets to interact with it. That way assistants can operate them too.
I have a Pixel also, and have issues with is since the change to Gemini. It works _most_ of the time, but every once in a while it'll tell me it can't set a reminder.
I use reminders often so I suppose it is a low failure rate.
But, when they first made the change to Gemini I had to switch back for a few weeks/months before it could set reminders properly.
I want the reminder not on my device, but on my work computer which I'll be staring at when I need the reminder. Or in the bananas case I want bananas sent to our shared family grocery app (which isn't a google service) so that if my wife happens to be going by a store tomorrow she knows to stop and get some along with flour or whatever else we need to stock up on.
I've set a reminder through Gemini, then tried to change its time. Told me there are no active reminders to change. Created another reminder with correct time. Got both reminders.
I said "Hey Google instead of reminding me about the bananas on Tuesday, remind me on Wednesday" and that worked. It's quite possible these things only work on Android 15 QPR 1, or only on a Pixel Pro, or who knows. Its capabilities have noticeably expanded in the last few months. One thing I know it still can't do, that the old Assistant did, is control a Sonos.
ETA: The joke is I don't even have Google Tasks installed, so I am not sure what effect those Tasks items would have. They might only surface on the side pane of Gmail.
Here, it is more like the smartphone market is stagnating and manufacturers need something to encourage people to buy new smartphones instead of keeping the one they already have.
They are already doing planned obsolescence, with hard to replace batteries, limited software support combined with closed systems, etc... But they can't abuse it, as regulatory agencies are already after them and customers are starting to notice. They had a go with cameras, as it is a significant differentiator, but nowadays, most smartphone cameras are as good as they can be for what they are used for.
And it turns out that AI is the big hype right now, so of course, they are using as a selling point.
The funny part is that "AI" (machine learning techniques) have been running in smartphones for a while (I'd say a decade), hidden in camera software. How do you think these tiny cameras can do pictures that look so good? But they mostly kept quiet about it, as it had implications of making "fakes". And yeah, stuff like Siri that works mostly on servers, you don't need a phone with "AI capabilities" for that, just internet access, but they are certainly going to put in in their ads.
I can understand why. Most of the generative AI crap we're being force-fed these days is a solution looking for a problem. On my Samsung, the only really useful AI features they provide are erasing things in photos and upscaling. The generation is weak compared to even basic stable diffusion and otherwise they're all fluff features and sometimes give me the same vibes as the "Make Longer" feature on Notion. As far as apps go, I really don't need a chatbot in every app.
It's incredible how bad it is. It actually seems to be getting worse year by year. I nearly crashed my car on the freeway trying to set a reminder to pay a toll the other day - I changed my mind and tried to get siri to cancel the action, and no matter what I said, she kept asking "what time? what time? what time?" on endless loop, and when siri is activated on my car's dashboard, I can't see the map - forcing me to avert my attention from the road, disconnect my phone, then plug it back in and pull my map up again.
She can't do or understand the most astoundingly basic stuff. I guess maybe it "feels" worse now because most LLM's are pretty good at understanding your meaning/intention, but my god, it's so bad that if I were in charge of that product I'd rip it out entirely. There's no way anyone finds any real use out of it.
Even if everything worked you should not be trying to do that while driving. Keep your attention on the road - you are in control of several tons of metal moving at deadly speeds. (don't sing a long with the radio either)
I'm unsure of your point. Are you arguing that issuing a verbal command to my voice assistant is not paying attention to the road? No more than talking to a passenger, really.
Yes. I'm arguing talking to a passenger is not something you should be doing either. You are in control of something deadly, not a video game or toy - act like it.
This seems unnecessarily snarky and way off topic, not to mention a bit silly. I've piloted much, much larger things professionally than a car, thanks for your concern though.
Had I not been flustered and in insane traffic I probably would have been able to figure this out on the spot quickly enough. I did eventually try this after a few minutes and it still wasn’t working. Alternatively, the product could just work how it’s advertised.
FWIW you were not forced to distract yourself from the road - officially, you should have got off the freeway at the next exit and parked before sorting stuff out.
When something is called CarPlay and actively marketed as a way to use your phone services whilst driving, you would expect it to have a good enough UX that doesnt detract from safe driving to use it.
I have a bunch of "premium" smart outlets(eve energy, some are "THREAD" enabled, all the bells and whistles) all named and connected to things I want to toggle with voice: a small radiator, an air purifier, a humidifier, some accent lamps, a fan, an infrared therapy lamp etc. -Before anyone asks, I ordered the European version of their energy strip for higher wattage stuff-. I give each one a clear and unambiguous name, and still, at least one in ten times Siri will be confused about what I'm asking and turn off EVERY SINGLE OUTLET IN THE ROOM. It's endlessly frustrating. For what it's worth the outlets never de-sync or disconnect like the random amazon ones do so I blame it purely on Siri.
I need better speech to text. I have to repeat my text message 5+ times sometimes when sending it through carplay. I'm not talking about anything long either. A handful of words that my accent just doesn't work with. More than once per day I get fed up enough to pull over and text it.
siri’s speech recognition and intent handling is so comically bad. as of late, for some unknown reason, when i ask siri to favourite the current song while driving it curtly replies that it doesn’t know which speaker i’m referring to. this used to work.
another fun problem is siri not recognising my speech despite me not having a particularly strong accent, speaking slowly, and enunciating. i’ve gotten into the habit putting on a valley girl or bbc news anchor voice while using siri since that usually works.
whoever is in charge of siri needs a reality check, the feature is borderline unusable.
It’s gotta be just barely holding on in some legacy environment right now. I’m surprised they don’t just start over using new tech. Maybe that’s the end goal with on-device inference to sunset cloud Siri.
I actively don’t want AI, on-device or not. But with the near monopolies on these devices there isn’t a way to vote with your wallet. We’re getting whether we want it or not.
The hard part with 2FA over SMS is that it's no longer considered secure [1]. I want a dumb phone too, but with all the security tools we need (password manager, 2FA apps/tokens, encrypted messaging, etc.) it's becoming less and less an option for me.
I wish there was a middle ground where I could have my phone be dumb enough to keep me from playing on it all the time, but secure enough that it makes sense for me.
I think we could presume an answer - if the respondents first received a full accounting of how their phones track and record their lives, along with a full list of who is getting that data.
Ok my next comeback - people are treating AI tools like a musical instrument.
It’s like picking up a guitar for the first time, twanging a few strings and saying ‘Nah, this sounds shit’. Guitars are useless. I’d never pay for a guitar.
Even chatGPT has a learning curve. I save myself hours or days per day using it for all sorts of things. Anyone who says they can’t find a use for AI is just lazy and hasn’t tried hard.
There are lots of people who could benefit massively from using AI. Who have no moral objection to it. But they just dismiss it so quickly because it doesn’t instantly, magically make beautiful things for them.
Anyone who cannot get something useful done by AI - who does want to do that - is lazy.
It’s like saying in the 19th century - oh this electricity isn’t really useful for anything much beyond lights. What’s the point?
The poll is very weirdly constructed. It looks like they gave people a set of checkboxes to check rather than asking people to rank their opinions on a scale.
25% checked the box saying they don't find AI tools helpful, but only 14% said they do. Which means 61% checked neither box.
45% are reluctant to pay, but only 6% said they were willing to. So again, 49% simply didn't say one way or the other.
Yeah, it's bad. Excluding the people who checked neither is probably better, but it's still not perfect because it leaves out the people who read both options and would have said "undecided" if it were a choice.
Those stats are wrong in another way - everything is about tradeoffs. If the AI is really good I can pay a price to use it - the price depends on how good it is. If the AI is really good I can ignore some privacy concerns - not all but some (indeed in order to work it probably must ignore some privacy concerns)
You can't say that unless you've seen the test. Was it an Yes/No question, "check relevant to you", or "write how you feel about AI features"? It would change responses drastically
“Turn off the local AI and let me have that extra power” . I think that’s what most people think. I certainly use AI for some tasks, but I don’t need AI everywhere 24/7 in my pocket.
Smartphones are an absolute graveyard of fads; remember the 3D screen phones, the phones with projectors, and so forth? They generally go nowhere. I suspect 'AI' on phones will be similar.
Overwhelmingly, what people want out of phones is "like my current phone, but with better battery life and maybe a better camera." Previously 'faster' was also a concern, but modern phones are largely Good Enough.
There's a lot we quietly take for granted. Supposedly the NPUs on phones were originally added to let the phone identify objects in your photo library. When shopping for phones, I tend to shop for better battery and camera... but I also wouldn't go back to a photo library without identification.
Yeah, I should possibly qualify; by ‘AI’ on phones I would expect that the average person polled understands LLMs, image generation, that sort of thing. OCR, image classification, etc, I doubt they’re thinking of that when answering this; that wouldn’t even have been marketed as ‘AI’ until a couple years ago.
My favorite one is Amazon's Fire Phone, which had multiple cameras so it could figure out where your face is relative to the screen, which would enable some crazy 3D effect.
Even camera improvements are overblown. My usage and photo quality did not improve much since ~iPhone 6. Taking decent travel photos, selfies, etc - I was happy with the results 10 years ago.
Technically, I understand the difference in the technology, I just don't know who needs that vs who gets excited about new features for a brief moment.
I would love to take a photo using my smartphone that doesn't look pixelated, blurry and or over-processed. Maybe asking too much considering smartphone sensors can't compete with DSLRs in some situations, but I'm always baffled with how dark and desaturated some of my photos turn out on my smartphone, as well.
I’m often surprised that my iPhone photo app show old photos in the widget that look stunning and were taken with the iPhone XS and 11 PM. Night photos are better now but day photos stalled a long time ago.
I find joy taking some shots with halide or photon in trueRaw where all processing is removed and grain is present. It gives an authentic look … so all HDR stuff is often too much
Few years ago there was usual 'Apple look' on Instagram where every photo was taken at golden hour, all skin blemishes and moles ironed out into oblivion. Everybody was doing it in some form, but Apple was (is?) going furthest, into territory of painting more than representing reality (I compared my Samsung ultra with wife's iphone and its consistently this... but generally all reviews I saw few years ago stated the same).
Now pendulum has swung so all go for 'realistic look', but I expect people actually want rather milder version of above.
Phone photos look OK on phones, but enlarged even the top contender from current dxomark show very much how hardware limits work. Its just not presentable, maybe apart from very bright scenes. Now I wouldn't go bashing phones per se, its marvel what they achieve from those tiny plastic lenses and some CPU time. And they are always there. But any low hanging fruit in phone photography was picked up long time ago by whole market, what lies ahead are slow computational improvements, coupled with very slow increase in size and thickness of camera section of phones to capture more light.
I think some people do genuinely care about the camera improvements; personally anything I’ve used from the iPhone 7 on has been _fine_, but I’m not a demanding camera-user.
Google photos search doesn't work well at all for me.
Even things like "[my dog name] beach" isn't reliable.
Or things like I use photos as notes. It doesn't reliably recall things like cheese when I take pictures of cheese in various stores to remember what is sold where. Not even the name of the cheese; just cheese. Ditto spices.
Don't use the name - "Mr Floppy at the beach" is meaningless, but "dog at a beach" will probably yield a lot more.
I've found google photos search to be pretty good, and if it can't find something usually the map-mode is enough to pin it down (e.g. go to the beach where you took the photo and it shows you the photos from there)
I did just check, and "dog at beach" generates sub 20% recall for me. I go to the beach weekly with my dog, take lots of photos because I'm a dork, and that first query skips many weeks.
Also, I did add my dog as a known / named pet under the explore tab, which is why I thought the name should work.
I can make it work by picking out the beach via geo, but I think the whole thing illustrates how much better this could be. I'd like to be able to get responses to queries like
* [pet name] at [beach X]
* [pet name] with sand on face
* dead seal, or even just dead animal (pics on beach)
* seaglass (recall is poor there too until I manually added to a photo album)
* dent in car
* [spice name] (I take pics of spices to know which stores offer what)
etc etc. The only way I manage the thousands of photos I have now is by carefully sorting into hundreds of albums, which google also doesn't support well.
Amongst the many many deficiencies of the app (which, tbf, does work extremely well as a read-through cache and seems to back things up very well), it likes to surface spotlights of dead people and pets. Which is not at all what I want proactively surfaced.
For it to have value they need to effectively apply it for something useful and have it be reliable, not just go hey we have llms. like if i could say siri share my location with my wife and text her "im on my way" oh and send her a pic of what im looking at through my sunglasses. And it worked 99% of the time, and could figure out to do it through google maps because she has android, that would be great.
And, to me, none of that even sounds like AI. Just an assistant app, smart glasses, and google maps not being shit. I can't think of a reason, other than the reasons I already use ChatGPT and the AI object eraser, that I would need AI on my phone. Most of the crap they pump out is just a solution looking for a problem.
Maybe it's just me, but I'd prefer to spend less of my life on my phone in general. Seems like first-party integration of AI features is just a ploy to persuade me to use it more, since the appeal of novel apps or software features has long since died out.
My Sonim XP3+ has a hard shell replaceable battery and lasts a week and a half on a charge. It's not even a very expensive device, it's under $150! ;) But it is a flip phone running Android Go with no apps...
That said, I think the tradeoffs being made right now are probably the right ones. Apple's latest devices have gone to an electrically released sort of adhesive (versus the older pull-strip removeable adhesive, which is a big step up from the "glue it in" approach many vendors take), and for a given volume, you get more battery if you can rely on the phone to protect it from damage - which is why almost everything with an internal battery uses some variety of pouch cell. They're quite a bit more fragile than the hard-cased batteries, but you get a lot more battery in the volume than you do with the hard cased ones.
As long as it's not incredibly irritating to replace the battery, I'm fine optimizing the daily use thing (battery life in a given phone size) over the once-every-few-years thing (replacing the battery).
Getting someone to replace a battery every 4 or 5 years really isn’t onerous. And for iPhones the magnetic batteries that attach to the back are a pretty effective way to stretch battery life a bunch more hours if you need to.
The phone manufacturers have to know this, right? they've got market surveys and focus groups and internal dogfooding programs. they all know that a half-baked chatbot experience is going to sell zero phones.
All this AI marketing push has got to be because they think investors are stupid, and they can fool the market into thinking they're doing an AI.
I’m all for AI—if it actually does something useful. But features like swiping to remove unwanted items or ensuring everyone smiles in a photo don’t count. For anything genuinely useful, I’d rather subscribe to an AI service separately than pay a premium to have it locked to my phone. I want access from my PC too.
AI is a means to an end - it's the end that people want (or don't want). For example people like fast and accurate spell check - they don't care whether AI did it or not.
It has been especially frustrating to read about Apple Intelligence on my 2022 iPhone with iOS Safari, where some JavaScript bugs that have been around since 2012 seem to be slightly worse in the latest update, and occasionally the entire app just freezes for no good reason, even if I've closed all other tabs. Not to
mention how often I have to reset my entire 2021 iPad because the audio drivers get into a bad state.
I suppose "stability improvements in WebKit" doesn't do much for Apple's stock price compared to SiriGPT. But this is feeling like death-by-1000-cuts: I don't think users are deeply committed to Apple UX/etc. I believe Apple's US market dominance is largely due to burnt fingers around Android's unreliability and annoyances between 2010-2020 (e.g. Google and Samsung not playing nice, badly orchestrated version changes). This was never a permanent state of affairs; Android has stabilized significantly and is more harmonized among competing manufacturers, while retaining its advantages on price and ease of development.
Apple seems complacent on the basics, and is overextending into an AI product few users seem to want.
Do buyers even know what features are there because of “AI”?
For example, on iOS, you can copy text out of photos (https://support.apple.com/en-ph/120004), you can search for photos without entering key words (not that well yet, in my experience, but the results are better than no results), predictive typing apparently uses a language model, modern camera apps do zillions of things to make photos look better, etc.
Neither of those are killer apps, but each does make the device a little bit better.
Everyone sees through a stupid lifestyle theater instantly.
Life isn’t just driving around taking filtered selfies and finding friends faces in albums.
If phones didn’t suck, advanced functions would already be there without “ai”. But they lack any sensible integrations beyond the marketed scenes where clueless people are playing particular job, live and hobby moments.
On iPhone, if I take a picture of a plant or animal, it identifies it for me. It's not 100% by any means, but it's useful enough. I've figured out what were baby plants I wanted vs. weeds. I've figure out species of birds I'd taken photos of with my SLR (ie: phone takes picture of Lightroom editing the image, and is able to identify it from that... I'd prefer there was a way to not require me to take a photo of my monitor, either doing it "live", and/or adding the functionality into the Mac.) For people and pets it can find other images that contain the same subject.
When my daughter was studying Chinese, I could use the live-video translation app and see the lesson text translated to English, and see her hand-written answers also translated to English. I could see this being more broadly useful when travelling, along with live translation of spoken words.
One use case could be improving navigation directions. Right now, map apps provide granular, step-by-step instructions that include unnecessary details, such as how to exit your own neighborhood.
AI could provide more human-oriented direction that focus on key landmarks and decisions rather than every minor turn. For example:
"Hop on 80 West, cross the bridge, take Sir Francis Drake onto 101 South, take the Alexander Avenue exit, don't go through the tunnel, and your destination will be on the right."
Well, I'm still waiting for an AI feature that recognizes my usage patterns, and adapts the system's behavior.
E.g. if it sees that I always reopen an application 2 seconds after the OS kills it in the background, then maybe it shouldn't be killed.
Or if I wake up 3 minutes before the alarm would go off, and take a trip to the toilet, maybe it shouldn't blow up the speaker while I'm frantically pulling up my underpants, but recognize that I'm already awake, or at least wait with the alarm until I'm around the phone again.
Or automatic backlight shouldn't go crazy when I walk in the night under the streetlamps, it should recognize that lamps are coming and going, and that backlight adjustment every 5 seconds is silly and annoying.
I could go on. IMO there is definitely a place for machine learning/AI in phones (and other places too), especially for quality of life thingies. Just nobody is doing them, I guess becacuse these are not as visible as image generation. My credit card has been ready to spend on such developments since at least 2021. One of these days I will have enough of waiting and do it myself, out of spite...
Spellcheck, voice control, voice-to-text, autocomplete and next-word-prediction are all some AI features that are already in use. Voice-to-text could certainly be much better if something like whisper was integrated. I pretty much never actually listen to voicemails, so having a reliable transcription there would be great.
I'd also love to be able to give commands that traverse multiple apps (e.g. take my google sheet and venmo request everyone the specified amount). Most likely this would happen by teaching an AI tool use and having apps expose an API.
I'd love to be able to give voice commands for certain things (e.g. flipping through recipes when my hands are wet) and have the phone be able to do the actual thing I want.
I actually think phones are a much better place for AI since they're so difficult to type on that voice could provide a higher-bandwidth interface.
I'd love it if CarPlay/Siri just could read out stuff it finds on the Internet. Currently, all I can get out of it is "Sorry, I cannot show this to you right now" for basically everything except trying to control multimedia.
At one point, I had ChatGPT working via voice in CarPlay mode (via Shortcuts I think?), but seems like Apple disabled that at one point, for some stupid reason probably.
That's not "on a phone," though. That's just schlepping an image up to the Google data center, and getting a result back. That you're using the phone as an interface to a datacenter doesn't make it "AI on the phone."
The only one I use regularly is object replacement in photos. It's great for editing a street sign out of a picture of the sky or something, especially if you just don't want to dox yourself posting a pic. It's definitely not high quality most times. Just blurry redraw of what the background might look like.
Otherwise, totally with you. No idea why my phone needs AI. I can just open the ChatGPT app if I want to have a discussion with ChatGPT about something. I'm so tired of apps updating to "Add a new AI assistant!" like why do I need to talk to an LLM in most of the apps I use?
This makes sense, AI is a new phenomena and most people aren't using it on their phones as much as they use SM apps, camera. Battery life is one of the major reasons why people upgrade their phone (believe it or not, many don't just replace their battery)
People think of chatGPT now when they hear AI, but for on-device its more like speech recognition, image keyword search, and keyboard next word prediction.
AI is a technology, not a product feature. I don't care about the underlying technology in my products, I care what they do for me. As long as the feature is great, I don't care if it's made with traditional algorithms, AI, or literal magic--it doesn't matter!
This idea that customers want AI is like saying that customers want applications written in Python. Why would they care what's behind the curtain?
I could not care less about AI on my phone. I don't want to learn to use it nor do I want to give it free reign over my content. If I can disable it on any smartphone I own I will.
Of course, aside from checking email and calendar I don't do a lot of work from my phone.
The only AI thing that excites me on my phone is a battery-friendly way for app developers to run ML models. I don't want an AI assistant that sees everything that i do, but some things like computer vision and translation and OCR are handy to have a super efficient coprocessor for, and I could see interesting apps using ML increasingly. But that's kind of a second order thing. I'm sure apple and them are also trying to figure out how to also let app devs use their strongest foundation models too without every app wanting to download 100GB of weights.
I calculated and found almost no correlation (r=0.2) between battery life and price. Which is quite sad considering it's the most important feature to consumers.
> The biggest motivation for US adult smartphone owners to upgrade their devices is longer battery life (61%)...
Battery replacements are a thing. They're apparently not a common thing, unfortunately, because that might impact new phone sales... and even on devices where it's trivial, people don't seem to do it. A decade ago or so I was doing a lot of Nexus 5 battery swaps for people because one of the battery OEMs was shipping junk and the batteries were shot in a year.
I really wish OEMs would put bigger batteries in phones. It improves everything. Not only do you get longer initial battery life, you can handle far more "battery aging" before things stop working right. You still have a day's battery life (which I expect is what most people actually mean - they want their phone to last the day without thinking about it) even with capacity loss, and a larger battery can have more internal resistance increase (another factor of battery aging) before it sags too badly under the load to keep voltage up.
Based on the fact that easily 90+% of phones I see in the wild have cases on them, physical size and thickness isn't a big factor (and those newer folding screen devices are massive when folded). Another few mm doesn't matter when phones live in purses, men's pockets, or jackets.
... or just go back to a modern flip phone sort of device, get a week and a half battery life, and stop worrying about remembering to charge your phone. ;)
> Based on the fact that easily 90+% of phones I see in the wild have cases on them, physical size and thickness isn't a big factor
Well, people are probably going to put cases on their phone regardless, so the phone has to be thin enough that even with a case it's still a manageable size.
But I agree that battery is where phones ultimately fail most of the time. I have had a Samsung S10 for a bit over 5 years now and it is the battery that is going to force me to retire it and get a new one. Speed, storage, screen quality, camera quality are all perfectly fine for my use case. But the battery rarely lasts a full day with any kind of usage now which is annoying.
I would love a phone with a replaceable battery. But I agree that the average consumer won't bother, they will just get a new one (or they will have upgraded before battery life ever becomes an issue).
I'm aware it can be done and have looked at the page on iFixit, and may end up doing just that. But as well as looking like a bit of a PITA to do it, I'm not particularly confident my phone will be good as new after I melt the glue and pry it open. My point is that it would be great if replacing the battery was as simple as clicking off (or even unscrewing) the case and taking it out, like it used to be. Granted, I probably couldn't take my phone swimming with me, but I think I could get over that.
IMHO the era of "we assume you're going to have a case on this" was cemented in place the instant the camera bump appeared. If you can't set you phone down on a service without it being all tippy thanks to the off center camera bump then you're clearly expected to stuff it into a case that is at least that thick.
I just did a battery replacement on a Pixel 6a. It wasn't terribly expensive, $18 for the whole kit, but the process was very nerve wracking. Unlike Apple's relatively friendly rubber cement Google used some kind of industrial adhesive to glue everything together. You need a good amount of heat to loosen it, but not too much because that will damage the phone. There's a fine line. They do include a helpful strap for removing the battery, but you have that awful glue to deal with. The cherry on top is the slippery little spring clip connector you need to remove without damaging the board underneath.
All in all it made me pine for the days when you could just pop the back off of the phone and yank the battery out to replace it. Personally I think I could live with a 0.8mm thicker phone that still has this feature. Also a headphone jack while you are at it.
> Based on the fact that easily 90+% of phones I see in the wild have cases on them, physical size and thickness isn't a big factor
Everyone says this, but I wonder if the sales data says otherwise, because device manufacturers, especially Apple, REALLY love to advertise how thin their devices are.
Personally, I don't want a super thin device because I feel like the slightest bending force is going to snap it in half. It's absolutely wild to me that people will store their phones in their back pocket.
Apple's mobile division still seems to struggle with the ghost of Jony Ive. At least their laptop division was willing to rethink things with the Apple Silicon transition and go back to things that worked.
The late Intel-era portable Macs were simply broken, because of Ive's apparent obsession with "thin for thin's sake." Or, more probably, "thin for the sake of industrial design awards." They innovated 20% too much on keyboard design with the butterfly keyboard that was simply a bad design. I don't know what else to call a keyboard that reliably breaks after a few years of service. But, bad though that it is, it was even worse because you couldn't replace the keyboard. You had to replace the whole top case and battery, at something like $700, when one key went bad, because the whole thing was integrated to save a mm or two. Not only did they un-solve a solved problem (reliable laptop keyboards), they even un-solved the ability to replace a keyboard when it gets mangled, or something spills in it, or a random failure happens.
But, worse, they went with the USB-C only design (the thinnest port!), meaning that every owner of one of those laptops I knew had a pile of assorted adapters in their bag so they could do things that most people want to do every now and then with a laptop, like drive an external display for a presentation, read a USB-A storage device, or read a micro SD card. Yes, the laptop was thinner, but at the cost of needing to carry extra stuff with you to work around the lack of ports.
And then, because it was so thin, they didn't have space to properly cool the zorching-fast Intel chips they put in them, and the chips would throttle badly while baking the machine.
The M-series laptops, shocking me incredibly, threw all that "design" out the window, and went back to something actually useful. They cool well (admittedly, that's mostly the Apple Silicon barely needing cooling), they fixed the keyboard, and the machines have a useful set of ports on the side for "common tasks with a laptop" (SD, HDMI, headphone, plus the USB-C/Thunderbolt ports - still no A ports, though). And then enough battery to last just about forever.
Anyway, I hope their phone division works this sort of thing out too.
Of course, they are gimmicks to attempt to sell new phones, now that the market has stagnated, many folks even have two or more phones, and hardly need fancy features.
I think the mistake a lot of companies are making with their approach to AI is making AI some "other" interface that you have to interact with explicitly. This chat bot approach is only one paradigm for using AI and it's probably the worst one in most use cases.
Amazon Rufus, for example, falls into this category and it's hard to even remember to engage with it. On the other hand, Amazon is happy to serve me absolute trash search results.
What most people probably want is an AI that can, for example, help organize their calendar transparently. Have a meeting coming up with a client? What if the AI can use tools to go find the latest info about the client? Check their LinkedIn, check their X, check their blog, research their company, give you a tear sheet on topics to discuss based on this in the calendar notes. If a user has to directly interact and instruct the AI, it feels like it defeats the purpose.
This can't be right. I've been hearing about how great AI is for the past year now. It couldn't be that media outlets just report on whatever they think will generate the most revenue.
Headsup: CNET is one of those sites playing videos (maybe relevant, maybe not) even on mobile. Be prepared to kill it if you're on metered plan (uBlock Origin can do that on Firefox for Android).
I still can't control my iPhone with my voice in the car. Siri doesn't work well for me. Maybe there is another way I have missed, but I'm amazed this functionality is still lacking.
A bit of a chicken and egg problem, isn’t it? Since really useful AI feature’s don’t really exist yet, why would one expect the general public to want it?
This is an underrated take. Apple, especially, wants to leverage their excellent silicon to do “AI” locally in the name of privacy purely as a marketing advantage, but the tradeoffs there are major (burning my battery instead of data center AC power, plus, my phone will never be better or faster than some big metal GPU with fans), and the privacy issue could just as well be solved by doing a good job designing the server side architecture and educating people that unlike when talking to a person, it’s trivial for a model to “not learn anything from” a conversation, and that that’s in fact automatically how it works unless you’re logging the conversations and purposely feeding them back in.
Building a privacy aware cloud is a lot harder than you think it is. Apple is going the whole way with remote attestation, hardware root of trust, software BOM, etc.
The problem is that remote solutions to the privacy problem inevitably involve my trusting someone else's privacy assurances. On-device processing reduces the trust required, although of course it doesn't eliminate it.
The common response to this is to point out that I'm trusting Apple or Google anyway just by using their phones, which is true, but (1) since phones are directly accessible to security researchers, there's more potential to find out about malfeasance if data is exfiltrated from the phone in a way that it shouldn't be, and (2) even if I have complete trust in Apple or Google, I also have to trust that they'll be able to protect my data from malicious actors, and, the less data I give them, the less I have to worry when they fail.
What surprises me about the story is how weak the effect is, which is stripped by the title mangler: only 25% of users report not finding AI features helpful.
The AI integration I'm looking for isn't to make chatgpt like queries or generate images or find one of my pictures, but to control my damn phone at a much more granular level. Settings are often getting removed or hidden or moved around to the point where controlling your device is a nightmare. Some examples of deeper configuration I'd love to see available:
- "Hey siri, stop interrupting my audiobooks and music with Teams notifications."
- "Could you please tone down the number of audio GPS alerts you send me? I don't need to be told five times before an exit that my exit is coming up. You're interrupting my audiobook too much."
- "Why are you interrupting my audiobook again?"
- "I don't know why soundless gifs keep stopping my audiobook, will you cut that shit out?"
- "Please let me use this audiobook app concurrently with this other app which has all sounds muted"
The foldable phone thing later in the article is fun, because I actually do want a foldable phone. I'd love to have a smaller phone with a screen that gets bigger for things like watching movies while traveling.
But the thing is, before buying one I borrowed one from a friend and it was absolute shit! The software hardly worked, the hinges felt horrible to open and it was still too big even when folded up. I thought I was going to break it constantly while holding it.
In re: AI on my phone - it's sort of interesting, but really not worth having because it's not smart enough nor capable of being smart enough. It lacks "judgement", which is what I really needs, instead of slavish attention to detail coupled with hallucinations.
I've been trying the iOS Beta 18.2 and here's my impression of Apple Intelligence:
It's not bad, it's really really bad.
The image clean-up feature is utterly useless, and I think it's one of the areas where we can clearly see the difference between a company for whom Generative AI is an afterthought (Apple), and the competition. I paid $1500 for the new iPhone Pro Max and a great part of the deal was the Apple Intelligence support, but frankly, I might as well switch to Android at this point because I'm really disappointed at Apple's take on AI. I'll probably wait until the official Apple Intelligence is introduced but tbh I don't think there will be much improvements over this version.
And as for Siri: It's as stupid as it ever was. I ask it to convert something from lbs to kg and it responds "there's no music playing". If anything, its natural language comprehension has degraded.
Currently, there's no context awareness, so I still can't ask Siri "how do I respond to this email?".
Really, the only thing that "works" is the ChatGPT feature that describes an image you send it. Anything Apple-related is bonkers. It's really embarrassing.
Apple has always sucked in AI and they have never formed the culture necessary to enable and attract top ML talent. They developed Siri years ago and have stagnated since.
As far as the future of compute goes, Meta seems like it has a much more compelling argument with the Orion glasses and their investment in AI.
They didn't even develop Siri, they acquired it. Doesn't seem like they've done all that much with it beyond integrating and marketing it, but I could be wrong.
With all the AI marketing for new phones, I keep yelling to myself "Nobody is asking for this!!" But maybe I'm in a minority group of what I consider "nobody".
Once someone manages to combine Siri's phone integration + ChatGPT Voice mode and it runs locally I will be very interested. Until then, I'm a bit meh on it.
I'd rather have flagship specs in a smaller package. The smallest in the iPhone (SE) and Pixel ("A"?) lines are still too big and tend to have previous-gen specs
I've been hearing this sentiment for 10+ years, but it's been tried and each time it's tried it doesn't sell well enough.
Because the execution is usually borked... I was eyeig ZenFone 9 (or something around that) and what? It was reported that it had problem with overheating and build quality.
What's more, I would love something akin to my current Galaxy a52s 5G with a display around 5.2-5.5" (I first had LG G2, then OnePlus3 which was already a bit bulky and now a52 as compromise; https://www.gsmarena.com/compare.php3?idPhone1=5543&idPhone2...)...
I do have iPhone SE (2022) and it's the size of LG G2 and I find quite handy. Something of that size but with slightly bigger screen (better screen-to-body ratio). Specs doesn't have super-hiper-premium… and the price should be sane (usually compact phones are like 20-40% higher, sic)
iphone minis weren't particularly borked, and they didn't sell well enough, people apparently preferred the larger iphones (to my own dismay)
In my case, they took so long to announce the iPhone 12 Mini that I gave up waiting and bought an SE even though it was slower than I wanted and had a poor camera. Four months later they announced the Mini, but I wasn't willing to replace a four-month-old phone. Then they discontinued the Mini line after 13.
When I was ready to buy a new phone, there were no iPhone Mini models for sale. It took more than a year, but I finally found an iPhone 13 Mini in stock on the Apple Refurbished store. Now I'm hoping to keep this phone alive until they finally release another small iPhone.
I'm using an iPhone 12 mini, and it's bigger than I'd ideally like.
Are you sure they bought different flavours because of the screen size or because of naming marketing "pro is better than regular" and "mini obviously has to be worse"?
Don’t produce it in the same volumes then?
Did you know that factories make less big-sized and small-sized shoes than average-sized ones? Because (surprise) buyers size distribution is not uniform.
“We tried to make big shoes many times and it doesn’t sell well enough”. Oh, really. I guess I’ll just cut holes for my fingers then.
Maybe the fixed costs of a shoe factory production line, in 2024, with centuries of production experience, are lower than those of a top of the line smartphone.
Speaking for myself, I 100% would have bought an iPhone Mini, but I purchase new phones on a 5-6 year cycle. My iPhones were the 4S and the XS.
Now that I'm ready to buy a new iPhone, the Mini has been discontinued! I think the Mini would have been and in fact was successful, but it's not successful "enough" to justify a separate model – they must have observed that people "like me" would still buy a flagship iPhone, even though we aren't 100% satisfied with the form factor.
Apple would rather have us buying a higher-margin flagship model and have an NPS of 65+ than a lower-margin mini model with an NPS of 80+.
My friend with similar instincts as me recently got a refurbished 13 Mini instead of the latest flagship. I'll probably get the flagship, because I value the satcom a little bit more than the form factor.
> I've been hearing this sentiment for 10+ years, but it's been tried and each time it's tried it doesn't sell well enough.
The iPhone mini was a billion dollar device. Anyone other than Apple would have called that a success.
That’s one of the big problems with these dominant trillion dollar corporations. For them it’s not worth to pursue something as small as a billion dollar market but they still have enough market power to suppress any competitor who wants to deliver such a product. That’s why we are seeing less and less innovation and diversity compared to the 1980s to around 2000
Just because the sentiment is a minority doesn't mean it's not an honest sentiment.
I also would love a smaller flagship spec'd phone.
I know it wont be popular, but I still want it.
Its the worst of both worlds. Not popular and harder to engineer. So it makes it hard to pitch to business.
It’s a package of devices Apple is selling. They sell under powered smaller screen macs. They sell iMacs. They sell small iPads. They sell smaller watches. Some of these just don’t sell as well as their other offerings.
They should sell small phones. Because the whole family will be in on the brand, features, and services. I’ve heard many family members and friends say that they won’t give up their older small phones because Apple no longer makes new ones.
The idea that they don’t sell well is not a good enough reason when you are trying to capture the whole market for not just hardware but the lock in for services, apps, games, music, etc.
How can you be confident in this when you don't even see their sales data?
Still, I would like to see a smaller regularly updated phone. Bonus points if there is a high-end version, because small shouldn't mean budget (like with iPhone SE).
The last small flagship (iPhone 13 mini) sold poorly, but it was much more expensive than the SE2. This was at the tail end of covid, but before faceid worked with masks, so the SE's touch id was a huge selling point.
Other than that and the camera, the only functional difference I can find are that the SE line is still missing the UWB antenna.
I'd happily upgrade to a newer small iPhone if they made one. As it is, it looks like the only option is repeatedly repairing my 13 mini (and dealing with the hilariously bad 5G battery drain forever) or downgrading to a newer SE3.
I know there are a lot of people in this boat. I predict they'll produce another small phone in a few years. It'll sell well due to pent up demand, and someone will be declared a genius for selling 100M's of extra phones that year.
> I predict they'll produce another small phone in a few years. It'll sell well due to pent up demand, and someone will be declared a genius for selling 100M's of extra phones that year.
I’d wish this too. I’m afraid that Apple over the next few year would become more risk averse then ever before. Also old Execs are leaving and retiring - so less people with hands on experience how to start new products VS keeping lights on.
They have repeatedly attempted to sell small phones. They don't sell in significant numbers, unfortunately.
"Well enough" is doing a lot of work here. It's not that they aren't successful, it's that they aren't immediate, runaway hits, so the manufacturers conclude: why bother trying to build this market, let's just go back to the playbook. That's how you get mediocre products, which is where we are now.
I'm sorry to say this, but again just face the reality of the fact that not enough people truly want this size of phone.
There's a congregation of people on HN who do, I suspect they're also the type of people who'll run their phones for 5+ years where their bigger phone buying brethren are replacing every year or two at most.
The market for the "small" smartphone just isn't profitable enough to bother with for most manufacturers.
I’ve got an appointment with Apple to replace the battery in my iPhone 13 mini.
I would love the new features (especially the camera and sat comm) but I’m not willing to get a bigger form factor device.
> I’ve got an appointment with Apple to replace the battery in my iPhone 13 mini.
I literally was in the Apple store yesterday for the same purpose (with a 12 mini). I'd also love the new features and hardware, but after trying all the available sizes in the store, they're all too big.
My wife on the other hand(s), loves to have a phone she needs to hold with two hands to even be able to use, so obviously she has the Pro Max. I don't understand how people are OK with that, but to each and their own...
To each their own, except apple's not making new mini form-factor iphones anymore so us mini preferrers will at some point not have our own.
there are dozens of us! dozens!!! i’m also keeping my 13 mini until it’s dead
I’ll upgrade the second a new mini comes out…but I’m keeping it for now.
I just replaced the battery in my iPhone SE 2nd gen last month. Somehow I didn't know that that was possible. Best 80 € spent ever.
I need a new battery in my 2020 SE as well, I wasn't aware that the last mini was a 13. Bring 'em back!
Good luck. I’ve tried to replace battery in mine 12 mini last week - with no success. I had to leave my phone for 4 hours or several days (if they brake screen during battery replacement, they will wait for replacement phone to be shipped overnight). Also representative was convincing me to buy a new phone - saying that battery replacement won’t help much because new ios versions has features which high battery usages, while newer iphones has larger battery and hardware optimizations for these new features. I’m thinking about iPhone 16 now while keeping iPhone 12 mini as backup phone.
So do the swap yourself.
Part and kit: https://www.ifixit.com/products/iphone-12-mini-replacement-b...
Steps: https://www.ifixit.com/Guide/iPhone+12+mini+Battery+Replacem...
I've successfully have replaced batteries and displays in older iphones (mainly iphone 6). But with newer iphones opening the phone is more complex.
I've read online and heard from Apple Store representative that iPhone 12 (all models) has tendency to crack the screen when phone is opened for repair or battery replacement and in that case Apple Store would replace the hole phone (this is were multi day repair process). So I would rather pay $90 to Apple that guarantees that I'll get a phone replacement in case when screen is broken during battery replacement. Without the phone I sill would be able to answer the cell calls from Apple Watch and with ipad over WiFi.
Oh boy I hope that doesn’t happen to me. It took them a week to get iPhone 13 mini battery in.
I recently brought an SE2 in for a screen + battery swap, which basically means they'd just give me a new SE2 at a steep discount.
They didn't try to upsell me at all, but I ended up getting an SE3 anyway (I didn't realize there even was a newer SE).
This is nice of them. I think this is because screen replacement was required and them not having SE2 in stock.
They would tell you and you would notice if your phone was replaced - it would have new serial and 2FA apps won't work on a new phone without reregistration.
Most likely it took them a week to get a new battery for replacement shipped.
Unfortunately, people don't buy small phones as much.
https://youtu.be/iR9zBsKELVs?si=3o1qD-4R7lyezZwt
There are a lot of different phones. People don't buy Sony as much as Samsung.
Smaller components cost more than bigger ones and the mini users (me included) also want to spend less, not more.
Same boat. 'upgraded' my iphone mini to a 15 and hate it. Far far to big.
My opinion is that most of the real uses of AI (like ML has always been) will be largely hidden things that are LLM based but not screaming at your face "AI". Particularly once the bubble pops and money stops being shoved into things just sticking an LLM in a pretty package with little to no value.
Some of the things coming in iOS like notification summaries and similar features are big examples. It's clearly LLM based but it's not a lot of the shoving AI needlessly into things that we are seeing now and provides a true improvement given the notification overload that we have right now.
Exactly. Ask customers "do you want AI in your phone?" and their response will probably be "meh", as shown in the article. But ask "do you want notification summaries, a better camera in low light, Siri to be able to look up more things, searchable photos, etc?" - and they absolutely will.
Photo manipulation is another example.
You can see the wisdom of how Apple is approaching this. In particular, to be on device whenever possible so as not to be dependent on network bandwidth, and to tie features to new hardware (to drive sales).
I have a hard time seeing how this isn't obvious. 95% of everyday AI needs (for the people that even bother to interact with it) are covered by ChatGPT, and most of that is the same stuff that Google was handling before.
From personal experience, the only thing that changed when replacing the "old" Google Assistant with the Gemini-powered one on my Pixel was that it's no longer able to create reminders.
The only thing that changed for me was a that it couldn't control my smart Home devices anymore, nor activate navigation, nor send messages via Whatsapp (I.e. while driving).
Literally every thing I used it for got answered via "I cannot do that yet" after it randomly opted me into that. Pure garbage.
Then why are you using it? I tried using Gemini once on my Pixel 6. Couldn't play music on Youtube music on verbal instructions, I switched back to Google Assistant. Will try it again after 6 months now. :)
Isn't that obvious? They tried to switch back to Google Assistant, but each time they asked Gemini, it said it can't do that yet!
> Then why are you using it?
I'm not, I pretty much just accepted that Google doesn't care about usability whatsoever and haven't prompted it in a very long time.
To be clear, the only time I've ever used it was via "ok Google" in contexts in which I'm unable to interface with the phone directly, i.e while driving. If it doesn't work you'll learn that you can't start driving before queueing the navigation anymore. The voice assistant was a nice feature, but not important enough to waste my time trying to figure out which feature they opted me into and how to get back out of it.
In my case it just kinda.. switched over at some point, and frankly I didn't care enough to figure out how I might switch it back (if I even could). I had a similar frustration to GP that it stopped working for 100% of the queries I used to use it for.
That said, at some point it started working better, but there was a good 6-12 months where it was a tire fire.
I'm not sure that's true. Not everybody is hounding for information from the web in the first place.
Apple's approach of using current-boom AI to help you navigate and digest your own private trove of multimedia content (photos, videos, apps, notes, structured data, etc) is absolutely useful to people as well, and for some of us, one of the only personal uses of this AI that seems compelling at all.
I'm much more excited to have help finding that goofy picture of my cat by describing what I remember, so I can share it with a friend, than I am to have some chatbot dialog about entry-level Python with a hallucinating parlor trick.
But these features have to work, and work well, and work fast, and be widely known to work, before they'll really win the market. But that's going to take a minute and it might not even happen.
>I'm much more excited to have help finding that goofy picture of my cat by describing what I remember, so I can share it with a friend, than I am to have some chatbot dialog about entry-level Python with a hallucinating parlor trick.
Hasn't Google been doing this forever? I can search random things in my photos (like pictures of an old car I owned).
For a bit, and to a degree, yes. Last-decade image recognition and tagging teased what might be possible and is genuinely useful.
The new LLM-ish tools promise that users can be more vague and casual in what language they use and more elaborate in how specific they mean to be; and that the queries (and operations) can span more diverse data sources.
Are there examples of new tools based on recent AI advancements that perform better than Google Photos image recognition?
Google highlighted the delta that recent advancements brought to their products:
https://blog.google/products/photos/ask-photos-google-io-202...
Wow they announced this back in May and it's still not available for me.
Or better than Picasa almost 10 years ago.
iOS has been doing this for a bit too. I don't use it enough to really know how good it is but I can definitely look for cats or people I know. Haven't used Apple Intelligence yet so maybe that's better as well?
Google photos is way, way, way better than apple photos at this. It’s not even a competition.
I have my sister’s dogs named in my google photos library. Every time I a take a picture of either dog, they are automatically tagged and added to a shared album I set up for my sister.
I have nieces and nephews with photos from newborn age to 10+ years old, and it has managed to organize them across their growth and ages. It’s incredible. I can search for “<niece name> <vacation area>” and get every photo of her on a certain vacation to make a family scrap book.
Apple photos search and tagging is pitiful in comparison.
But one of the two you can have on your device and do not have to pay rent
Yeah, it seems based on the advertising from the various AI vendors, they are showing its use by summarizing emails/phone call/etc. Things like being able to search text messages for info blah blah. The only one I've seen pushing online searches is Google, but that seems like duh! for them to be pushing. Circle something in an image and take me to a listing of that something for sale. Of course that's Google's direction.
But that whole find me something on my mutliple gigabytes of storage on my device account definitely seems like the mass appeal
>But that whole find me something on my mutliple gigabytes of storage on my device account definitely seems like the mass appeal
This is such a mundane use of AI, but unsurprising Apple would sell it as revolutionary.
Mundane or not, it’s actually useful, a meaningful improvement, and should work consistently well.
That is in contrast to a lot of fancy AI demos which are a great party trick, but fall apart in actual usage, with their reliability being “maybe it will work this time, maybe it won’t, just keep retrying :)”.
Apple is pretty well-recognized for usually being a bit late to the party, but at least delivering stuff that’s polished.
Just look at this thread of people sharing how Gemini broke all their commands and automations. The Apple Intelligence Siri on the other hand works just fine (even if new features are arriving slowly).
It's great for spooks too. Now they just have to exfiltrate the keywords describing the images instead of the images themselves.
I'm not sure that a hallucinated image of something is better than the original image when you're doing spywork.
The difference between whether someone has 4 or 5 fingers, whether they're holding a gun versus a random object, or whether they're mixed-race or caucasian, all seem like they would be pretty important things. Likewise, car number plates, signage in the photo that help identify where it is, metadata of the image itself (often more useful than the image), are all incredibly important. All of those are things that AI is absolutely terrible at lmao.
He's not talking about generating images, he's talking about classifying existing ones.
The problem is, vast majority of smartphone usage is done for entertainment and social networking purposes (IG, TikTok, Twitter, HN, gaming, Netflix and etc.). If you’re mostly scrolling and consuming, I can’t imagine how current AI tooling can help you other than some summarization of texts. Sure, for productivity cases it might be legitimate, but that’s not what supermajority of people use a phone for.
> If you’re mostly scrolling and consuming
Imagine an AI that popped up when you are reading something and warned you that information is false.
For instance, imagine something like https://theconversation.com/can-ai-talk-us-out-of-conspiracy... helping people discern about news and propaganda.
Why would you trust AI, something that regularly makes stuff up, to be able to accurately determine that?
the fun thing about the chatbots regularly making stuff up is that they almost always know when they're making stuff up. the hallucination problem isn't a problem of not knowing the facts, it's a problem of not knowing whether you want an accurate answer or a creative answer.
try asking chatGPT to only give you true and accurate answers and not make anything up.
I trust a defective AI a lot more than I trust Fox News ;-)
Now, more seriously, it'd need to put together a coherent argument and back it up with reputable sources, as just citing sources is very ineffective. The article I cited gives more details on possible approaches to that.
People eternally hoping that some new trick will finally make the other people understand how their side of the story/argument is one and only truly truthful. I guess this is old as mankind.
With your argument, the problem happens when given person goes to Fox news in the first place. Selection of the source has already been made, with its biases. Not much you can do or expect after this point.
Also, who curates the curator? Again an age old problem with no real, long term working solution in sight. No, you should not expect some statistical model to hold your hand through vast internet, while giving up any form of critical thinking, reasoning, or I guess any cerebral process altogether. Ultimate laziness. Since we know how much money there is in diet fad business, its safe to say this above will find its non-tiny desperate crowd.
> their side of the story/argument is one and only truly truthful.
There are no sides in objective reality. You might offer competing hypotheses and evidence for those, but we don’t need to do that to know that climate change is real, that it’s caused by humans, and that vaccines work.
That’d be useful but I’m pretty sure they’d get a massive backlash on Fox News and lawsuits filed alleging “being cancelled” within minutes of that shipping. It’s something we need but our current disinformation problem isn’t an accident but the result of decades of investment.
The less fraught one is warning users that they’re being scammed: https://arstechnica.com/gadgets/2024/05/google-wants-ai-to-l...
I can't imagine that working, even if that AI were more reliable than currently possible. Someone who already disbelieves credible reporting and objective sources is not going to be swayed by an AI telling them to disregard their fantastical conspiracy theories. Especially not in a near-future world where everyone is inundated with AI fakery.
That's even without considering the efforts that would be made to undermine any system that showed any effectiveness. A lot of misinformation, probably most of the stuff that gets traction, isn't random, it's serving a purpose and being pushed for that reason.
And who's going to teach the AI what's a conspiracy and what's not? Battles have been fought over what narrative certain Wikipedia pages should push; it's pretty hard to find information that can be objectively considered the truth.
Im not sure people really know how to use an AI chat bot or understand its a more powerful, quicker and personally more fun way of getting information then just a Google search
Some personal examples of how I find it more useful (love to hear yours) and fun to use....
- Wanted to go on a hike an hour away from where both my friend (lives two hours west of me) and I live. Asked GPT what are some good hikes an hour drive away from both of us to meet & hike. With Google I have to do Many searches where GPT just provides the answer right away.
- I count calories and eat out everyday. GPT knows the calories of everything i eat as I eat at chains mostly (Cava, Panera, Starbucks, Chipolte). I tell it via voice what i just ate for my 1st meal, it calculates my calorie count and later I'll tell it what im having for my 2nd meal. It can also recall my calorie count from days ago. It does all this quickly vs. Google i'd have to do oodles of searches.
Usually Im using GPT the most when driving via voice and unlike Siri, GPT understands me and i can have whole conversations with it to get things done while driving.
> Im not sure people really know how to use an AI chat bot or understand its a more powerful, quicker and personally more fun way of getting information then just a Google search
I can't speak to your personal entertainment experience, but AI chatbots are generally a slower, less accurate way of getting information than a google search. (Though Google polluting search results with a big, often inaccurate, AI result at the top narrows this a bit.)
If you use it for research where you have to do many google searches vs. just ask one question like hiking question it's much quicker asking one question vs. multiple google searches to get ur answer.
The thing is it isn't reliable enough to rely on the answer from just one question for anything that matters.
I don't really understand why it is acceptable to speak for others on this topic. It is fine if it doesn't work well for you. It is also fine if it works well for others.
These blanket statements lead to flame wars.
All of that sounds like the boring low-hanging life fruit that gets trotted out in videos by companies like Apple and Google as being "revolutionary." It's boring. It's staged. It's the easy stuff. It's well-off 20-somethings solving non-problems for other well-off 20-somethings.
Wake me up when I can say things like…
Hey, Google, are my custom license plates ready for pick up at the tax office?
Hey, Siri, ask my doctor to refill this medicine.
Hey, Alexa, how many charging stations are broken at the gas station on 16th street?
Hey, Google, why is this plant dying?
Hey, Siri, why are there so many people in my neighborhood today?
Hey, Alexa, did anything ever get done about that story in the newspaper from a couple of years ago about the Chinese slave labor being used to grow pot on illegal farms on the Navajo reservation?
"AI" just doesn't have access to the information required to do anything interesting or useful. And because so much of its information comes from the web, which is already so polluted on certain subject (gardening, travel) as to be useless, the AI becomes useless.
No, the problem is that "AI" just isn't any good at almost anything.
Changing to Gemini broke all of my smart home commands. Keeps trying to search/LLM how to turn off kitchen lights.
I've heard horror stories and have held off so far in 'upgrading'. In the end I don't really want the fully flexible responses people are leaning into with these llm tools. All I want is to be able to give a precise instruction with my voice and have the machine reliably perform the action that it performed the last time I gave that instruction.
Since that seems to be an increasingly niche desire (at least as far as the product managers are concerned), I've been looking more and more seriously at setting up my own local voice assistant. My main barrier has been hardware—the mic arrays in the Home devices are surprisingly good and hard to beat with cheap off-the-shelf components, and you need a good mic for good STT.
Yeah, I would like Siri to actually play the album I asked for and not something completely different phonetically from what I asked. Or even when I set an alarm and not be told "I can't connect to the internet right now" while I'm using my laptop connected to the internet. Or if my internet is down to actually use the speakers that I bought as speakers.
The hardware is really well done but the software is either over or under-engineered to a stupid degree.
It also sometimes asks to unlock my phone for commands that plain old Assistant was happy to do while locked. I haven't really found it useful at all yet, free ChatGPT is just better than free Gemini for "LLM stuff" and Google Assistant is better for "smart home stuff"
I've been thinking about trying the OpenAI integration for home assistant[1], because controlling things in my home is primarily what I use my assistant shortcut for. The normal assistant works well enough but can be frustrating if you don't remember the exact phrasing it wants to activate a certain command.
[1]: https://www.home-assistant.io/integrations/openai_conversati...
I have it set up with ollama. It’s… interesting. HA commands are provided to the model as tools so it works as well as the model is able to determine when and how to use tools. From experimenting with that and my own tool use code I’ve found that models vary greatly in their ability to wield tools and none that I’ve tried are exceptional.
It’s neat that you can intermix general chatting with HA commands but you’re probably going to find that the old assist is more reliable for commands. What I do like is that you can use a template as your system prompt so you can provide the state of a number of entities and then ask for them with natural language. That works well.
I have an Alexa/Echo voice announcement system set up and have recently tied that into assist so I can do automations like if the garage opens I prompt for “what is the state of the garage?” and announce the result. Makes it feel more humane than the same plain announcements all the time.
Do try it. I've been running it ever since it got integrated into the core, mostly to control A/C units around our flat, and it's the best voice assistant experience I had to date.
I mean honestly, how is it possible Amazon, Apple, Google and Microsoft[0] all keep screwing this up for over a decade now? I literally spent 15 minutes hooking up GPT-4 to the Home Assistant integration, and I was able to semi-reliably[1] control actual devices[2] like air conditioners and smart lights, in a completely natural and ad-hoc way, by talking to my smartwatch on the go, or to a phone, whatever was more convenient at the moment.
It's a really magical experience, a step closer to Star Trek reality. And what makes it possible is not just LLMs being able to deal with natural language, but more importantly, "bring your own API key" model allowing to cut away all the bullshit that FAANG assistants are stuck in.
--
[0] - Ever since they dropped MS Speech API in Windows, and did the Cortana thing. Some 15 years passed, at this point, and I'd still prefer to work with the Speech API than to touch any of the FAANGs' voice assistant - it worked, and worked off-line!
[1] - Works ~90% of the time; some 5% of the time the voice model (from Home Assistant Cloud) misunderstands me, and 5% of the time the LLM gets confused. It's still worth it, because I can actually talk to it like to a person, without thinking of style or grammar or magic keywords.
[2] - Which, given the level of integration of Home Assistant companion app with the phone, can be easily turned into an equivalent of on-phone voice assistant that can do more than the one I got from Google. Critically, there are ways to couple Home Assistant app and Tasker, so it's not hard to make it do arbitrary things on your phone. And, if you don't like low-ish code Tasker experience, you can trivially shell out from Tasker to Termux, at which point sky is the limit. Point being, an enthusiastic non-developer with minimal tech aptitude can beat Google and Apple at the voice assistant game today.
> Keeps trying to search/LLM how to turn off kitchen lights.
Teach a man to fish!
AI’s appeal depends a lot on the features. Battery is important for me (more than being anorexically thin), but I would love an AI that could screen my calls like a smart voice mail that asks questions.
Also, being able to talk to my mailbox asking questions about subjects mentioned in my e-mails would be a huge time saver.
Imagine a purely local Microsoft Recall-like thing that could answer questions about things you saw, or that read the news articles you went over quickly and answer complicated questions about them much later, at a time you just started to regret not having bookmarked it for future reading.
Computational photography uses AI a lot (next to parametric approaches), and without these algorithms, smartphone cameras would be quite shitty. People do care about that. AI isn't just LLM chatbots.
I know that's true, but I find that the images on my Pixel are starting to have a bit of an eery feel, with some of the details looking more and more like AI generated images. I'd give back a bit of the quality for more "natural" looking images.
That’s the line of thinking behind Halide’s recent “process zero” feature on iOS
https://www.theverge.com/2024/8/14/24220178/halide-camera-ap...
Here you go: https://www.theverge.com/c/24208960/digicam-digital-camera-c...
How do I disable that?
I'm having a horrifying realization that all of my pictures are "fake" in the sense that they don't match what I saw/experienced. Maybe it's time to get back into Polaroids.
> I'm having a horrifying realization that all of my pictures are "fake" in the sense that they don't match what I saw/experienced.
I'd caution that judicious/proper post-processing is actually needed if you want that result, because of the differences between the sensors.
Your human experience comes from many small pictures taken by a set of lenses panning across multiple points in a scene with constantly adjusting exposure times and focal lengths, all biologically composited into what feels like a single moment.
Trying to fully replicate that with a single artificial picture is going to be deficient in certain ways.
---
Separately, a pet peeve of mine: Too many people have been subtly brainwashed into conflating the "like I was really there" with " like a Hollywood film camera was really there." Then the next thing you know your medieval fantasy game has lens flares in it for no good reason.
I recently bought a real camera partially for this reason. I don't even mind how inconvenient it can be at times because honestly taking photos with a full camera rather than your phone is fun. While on the phone it has become quite dull in my opinion.
But knowing that the phone does a lot of software tweaking to get a picture to look similar to how good a full camera is made me want to switch. I think this was around the time that article about Samsung basically replacing a photo of the moon.
What does this mean? Sure, if generative AI is filling in features that weren't present, anyone would call that doctored or fake. But computational photography is mostly about recognizing patterns and filling in assumed detail... which is also how your visual cortex works.
Get raw, and marvel at its imperfection and ugliness in many aspects. If given phone ain't giving up raw data, take any decent camera, raw sensor data has been part of it since its beginning.
But those pics will be probably further from perceived reality than those enhanced by software (lets not get retarded here and don't brand every data processing as 'AI'). Distortions not only of barrel type, waning brightness towards edges, moire, heavy vignetting, tons of noise, over/underexposition, maybe some dead pixels... thats not how I see my days go by.
What makes a Polaroid any more "real" than an iPhone picture to you? Can any photo truly be real? (Deleuze has some interesting thoughts on the matter)
Right but in street surveys nobody knows that. Most people just call it "better camera."
... then one day the tentacles come out ... https://i.redd.it/j1cr7pr6m7j71.jpg
Interpretation of this on its nose suggest the algorithms are the core feature not AI, as there is no artificial intelligence involved in these processes that I’m aware of.
If you actually peak under the hood they just pass through weighted selectors, no different than a switch statement
That's what AI is. The weights are the clever part. Cameras have done this since the Nikon FA in 1983.
what about the infamous recognizing the moon in the background that is just an over exposed white fuzzy circle and replacing with a stock image with full surface details? clearly there's some sort of ML/recognition of content within the image
Yes that would be ML as far as I’m aware, but the thing in reference is software compensation for image quality. Effectively smartphones automatically upscale photos by default. I think the only exception is when you choose to shoot in raw
>Computational photography uses AI a lot (next to parametric approaches), and without these algorithms, smartphone cameras would be quite shitty
Nah.
Smartphone cameras stopped being shitty a while ago, long before AI and computational photography hacks.
What you mean to say is they without AI, you'd know sooner that the smartphone maker put a cheap, shitty camera in your "premium" phone.
>People do care about that. AI isn't just LLM chatbots.
Yeah, it's also fake image generation featuring humans with a funny number of fingers.
What AI isn't is a camera.
> 95% of everyday AI needs
This assumes that the capabilities and use cases are unchanged. Yes, for the AI features available today, I suppose ChatGPT can do much of it -- I wouldn't know because it's not interesting or useful to me, so I don't use it.
But: If I'm deciding whether AI features are important to me in making a decision to spend money on a future phone, it's those future AI features that I will be assessing.
95% of my everyday needs for an external intelligence (besides my own) are covered by e-mail, text, and phone calls with other humans, with a trivial portion covered by nascent AI features. As this changes, and AI gets more capable of replacing human intelligence in these interactions (TBD if this happens in the next smartphone generation, or the next human generation, or further in the future), then I will /very much/ care that the electronic device that I use most often day-to-day has access to these capabilities, and will very much use access to those capabilities as part of deciding where to spend my money.
You seriously assess "future AI features" when buying a phone? Have you heard the expression "a bird in the hand is worth two in the bush"? Also, what is the lifetime of a phone? How far in the future are these new features expected?
No, I assess current AI features when buying a phone, in the future. Any survey about what features customers value in phones they are considering buying inherently ask questions about future behavior, occurring in the future’s present.
Apple will just tell you you need an iPhone 17 because it has a more special "you're gonna love it" neural thingy onboard, and so your purchase of a 16 is null.
> 95% of everyday AI needs (for the people that even bother to interact with it) are covered by ChatGPT
True, but it is an external service with all the privacy concerns that entails. I appreciate eg. Apple pushing local AI but at the same time I don’t think it needs to ship with the OS. Just provide an AI API apps can hook into then I can decide which models I want and where they run.
> the only thing that changed when replacing the "old" Google Assistant with the Gemini-powered one on my Pixel was that it's no longer able to create reminders
Weird! Mine can still create reminders, as well as set timers and alarms.
AI is usually about implementation details. Customers typically worry about solutions to problems. They don't care how solution is implemented - using AI or not.
Perhaps a step further: I do not need my phone to tell me an LLM-generated bedtime story involving a specified cast of characters, I just want it to reliably control existing functionality, dangit.
Like when a timed alarm is making the phone buzz in my pocket while I'm driving, I'm telling it to silence the alarm, and the response-voice is regretfully informing me that there are no alarms going off right now.
"Everyday AI" hasn't even been built yet tbh. Where is my assistant that notices my flight has been delayed and reschedules my appointments and lets my contacts know?
I think your example highlights why this kind of "everyday AI" would be nearly impossible. If my phone automatically rescheduled my appointments and notified my contacts, without my explicit direction, I would be pissed as all hell. There are some "confirmation" notifications that could assist here ("We noticed your flight is delayed - would you like to reschedule these appointments?"), but even then, I'd say the majority of notifications I get these days are annoying and I spend a ton of time trying to figure out how to turn off annoying notifications while not totally silencing ones I depend upon. I'd have a difficult time believing that AI systems wouldn't just add to my list of over-burdensome notifications.
If you were an important CEO or something, and you had a competent human Personal Assistant - would you allow them to reschedule your appointments if you were running late?
Seems to me the problems are (1) the "assistants" aren't anywhere near good enough to be trusted to make the right decisions, and (2) a trustworthy assistant isn't compatible with the adtech business model, so it's unlikely facebook or google would produce such a thing.
>If you were an important CEO or something, and you had a competent human Personal Assistant - would you allow them to reschedule your appointments if you were running late?
Yes, why wouldn’t I? I could also give them parameters like “if I’m more than 20 minutes late please re-schedule this” or “if my flight is delayed please let everyone know it’s delayed”
Why wouldn’t I do that? Presumably the person hired is competent to make determinations within parameters specified.
I could also let them know when it’s inappropriate to do this. Again, they should be competent enough to discern the differences between when it is and isn’t appropriate.
This could honestly be done by an algorithm if you give it the correct inputs and outputs and it could be fed updates, the only real limit is the fact that some of this isn’t exposed via an API either in a timely fashion or at all
Important CEOs bring their trusted human assistant with them. Those of us at a lower level don't get that option, but at the CEO level you get to hire your own sectary and you spend years teaching them how you work and what to do in all the situations that come up and so they will make the right decisions often enough that you can trust them. Or they know they are not trusted to reschedule that appointment and so they don't.
That years of training is what we are missing. I don't think modern AIs can be trained in the way the assistants of old could be, at least not yet.
I don’t even think you need machine learning for this, it’s an API problem mostly like I said.
Being able to collate the requisite inputs from outside sources is the real problem. If you can’t do that reliably it’s simply hard to build an algorithm around it. Flights for example would require your calendar program to reliably pull data from an API regarding the flight information that is current and effectively real time. That’s the actual hard part, and this expands across services.
For all the advances we have made with computers and smartphones in particular they suck at meaningfully exposing a way to collate data sources and create actions around them reliably
Personalizing any automation like that is a privacy nightmare. The system needs to know a lot about your preferences and the decisions you'd normally make yourself as well as your current circumstances since those will also influence your choices. How do you feed that into any AI without being problematic for privacy?
Having one running locally helps but it's still necessarily storing information that you might not want to have stored where someone could potentially retrieve it, either via some sort of exploit or by forcibly compelling you to give it up.
If I were a somebody worth a personal assistant/secretary, I'd be pissed about them doing such things without notice too
If you had such a person they would know if you wanted that service of not and act accordingly.
I worked on a (failed) system that was supposed to help with this in the pre-ChatGPT era. The obvious limitations is getting the data, and scaling that to everyone. Learning about your flights, staying up to date on them, etc is such a daunting privacy-busting task that everyone is scared to start. Either you go use-case by use-case (flights, restaurants, calendars… etc) and never get traction or you just start scanning emails and open up huge data risks.
Today, this is nearly available, nearly. Probably only something Google/apple can realistically offer. Apple “intelligence” has started to read your notifications and rewrite them for you, so it shouldn’t be a big leap to listen for a United App notification and decide it’s urgent enough take action. Should be “trivial” for Google to do as well, and they could even run it server side to help without a phone present.
That's agentic behaviour, and the first to offer it publicly are anthropic, who just opened a beta.
It's still pretty terrible though
I love how the goal post is being moved publicly as to what AI means.
AI is anything automated it seems, and now they’re being subcategorized into niches as to what they do, e.g. “Agentic AI”, “LLM backed AI systems” etc.
If it’s not real intelligence then it isn’t really AI, and I wish the world at large would call it out.
LLM, Machine Learning, Neural Networks etc are all great but none of them have true spontaneous intelligence or learning ability.
Please, someone point out how any of these systems have organic spontaneous learning ability for a subject it was not pre-data seeded on. This is a generally accepted measure of higher level sentience as far as I’m aware
> it’s not real intelligence
Hence the predicated “artificial”, and hence the downvotes you are currently receiving.
Your message is largely, if not entirely, a strawman.
None of these exhibit any accepted definitions of intelligence markers. The intelligence isn’t artificial, it doesn’t exist is my point. If you apply the commonly accepted definitions of what is considered display of intelligent behavior. One aspect of which is the ability to adapt to new circumstances that you haven’t experienced before.
There has been considerable success in programming computers to draw inferences, for example, but not actual reasoning. You can mimic some forms of reasoning but you can’t take one ML set - like recognizing photos with mountains, then expect it to correctly identify a similar geographical element - a hill. It can’t do that. It may correctly identify that it’s not a mountain but that isn’t the same thing as actually learning it’s similar to a mountain but not the same, which would be a rudimentary definition of a hill that an intelligent entity could conceivably use if it knew what a mountain was but not a hill.
Machine Learning was always a more honest place to have This discourse. I am indeed pushing back on the idea that we should be calling ChatGPT or anything like it intelligence.
It’s Machine Learning, clever algorithms, Large language Models, among other things, that are trained on ways to mimic certain aspects of intelligence, but it does not actually possess any real intelligence. Look at the LLM hallucination problem for example. It can’t be self corrected because it’s not an intelligent system.
Moving the goal post on what AI means (and pushing AGI as some new goalpost) is disingenuous, and relatively recent.
I’d care not if it wasn’t for the fact there is so much misinformation around capabilities and the future of AI, that it’s already negatively crept into policy making for example.
1. How does it know how/when to reschedule that doctor's appointment that you need to call to reschedule? How do you know that receptionist won't hang up on your "AI agent" who tries to call them, b/c they think it's some sort of scam bot?
2. How does it know which contacts to contact? Does that acquaintance you talked to for some professional reason need to know your flight got rescheduled? What about that travel agency you talked to last night to confirm the flight?
I would say the problem there is the requirement to phone a human to change a record in their database.
If they had an app then an AI assistant should be able to tie things together. Where things seem to be going is apps provide an intent-based API wrapper plus UI widgets to interact with it. That way assistants can operate them too.
I just asked my Pixel to remind me next Tuesday to pick the bananas, and it set that in Google Tasks. Which part of this doesn't work?
I have a Pixel also, and have issues with is since the change to Gemini. It works _most_ of the time, but every once in a while it'll tell me it can't set a reminder.
I use reminders often so I suppose it is a low failure rate.
But, when they first made the change to Gemini I had to switch back for a few weeks/months before it could set reminders properly.
I want the reminder not on my device, but on my work computer which I'll be staring at when I need the reminder. Or in the bananas case I want bananas sent to our shared family grocery app (which isn't a google service) so that if my wife happens to be going by a store tomorrow she knows to stop and get some along with flour or whatever else we need to stock up on.
I've set a reminder through Gemini, then tried to change its time. Told me there are no active reminders to change. Created another reminder with correct time. Got both reminders.
Failing to set reminders/alarms is core Google technology, #2 only to PageRank itself.
I said "Hey Google instead of reminding me about the bananas on Tuesday, remind me on Wednesday" and that worked. It's quite possible these things only work on Android 15 QPR 1, or only on a Pixel Pro, or who knows. Its capabilities have noticeably expanded in the last few months. One thing I know it still can't do, that the old Assistant did, is control a Sonos.
ETA: The joke is I don't even have Google Tasks installed, so I am not sure what effect those Tasks items would have. They might only surface on the side pane of Gmail.
I asked if to create task for Monday morning. It created one, Monday 12AM.
AI companies are desperately trying to find actual use cases but it seems like there is not that much to justify current investment.
Here, it is more like the smartphone market is stagnating and manufacturers need something to encourage people to buy new smartphones instead of keeping the one they already have.
They are already doing planned obsolescence, with hard to replace batteries, limited software support combined with closed systems, etc... But they can't abuse it, as regulatory agencies are already after them and customers are starting to notice. They had a go with cameras, as it is a significant differentiator, but nowadays, most smartphone cameras are as good as they can be for what they are used for.
And it turns out that AI is the big hype right now, so of course, they are using as a selling point.
The funny part is that "AI" (machine learning techniques) have been running in smartphones for a while (I'd say a decade), hidden in camera software. How do you think these tiny cameras can do pictures that look so good? But they mostly kept quiet about it, as it had implications of making "fakes". And yeah, stuff like Siri that works mostly on servers, you don't need a phone with "AI capabilities" for that, just internet access, but they are certainly going to put in in their ads.
It’s been enough time now that we should really have a handful of very clear use cases, but we just don’t.
Sounds a little like crypto although there are actual use cases for AI. Just not as many as investors think.
It's very similar in regards to treating the increase in compute power consumption as a signal of "growth".
There are lots and lots of use cases for good AI.
Those LLMs we have around just aren't that.
bingo
I can understand why. Most of the generative AI crap we're being force-fed these days is a solution looking for a problem. On my Samsung, the only really useful AI features they provide are erasing things in photos and upscaling. The generation is weak compared to even basic stable diffusion and otherwise they're all fluff features and sometimes give me the same vibes as the "Make Longer" feature on Notion. As far as apps go, I really don't need a chatbot in every app.
Never mind AI, I just want Siri to have better text-to-speech quality than my freaking vacuum cleaner.
It's incredible how bad it is. It actually seems to be getting worse year by year. I nearly crashed my car on the freeway trying to set a reminder to pay a toll the other day - I changed my mind and tried to get siri to cancel the action, and no matter what I said, she kept asking "what time? what time? what time?" on endless loop, and when siri is activated on my car's dashboard, I can't see the map - forcing me to avert my attention from the road, disconnect my phone, then plug it back in and pull my map up again.
She can't do or understand the most astoundingly basic stuff. I guess maybe it "feels" worse now because most LLM's are pretty good at understanding your meaning/intention, but my god, it's so bad that if I were in charge of that product I'd rip it out entirely. There's no way anyone finds any real use out of it.
Even if everything worked you should not be trying to do that while driving. Keep your attention on the road - you are in control of several tons of metal moving at deadly speeds. (don't sing a long with the radio either)
I'm unsure of your point. Are you arguing that issuing a verbal command to my voice assistant is not paying attention to the road? No more than talking to a passenger, really.
Yes. I'm arguing talking to a passenger is not something you should be doing either. You are in control of something deadly, not a video game or toy - act like it.
This seems unnecessarily snarky and way off topic, not to mention a bit silly. I've piloted much, much larger things professionally than a car, thanks for your concern though.
You could have just given it a time and canceled it later when it was safe to. No need to endanger others on the road for this.
Had I not been flustered and in insane traffic I probably would have been able to figure this out on the spot quickly enough. I did eventually try this after a few minutes and it still wasn’t working. Alternatively, the product could just work how it’s advertised.
FWIW you were not forced to distract yourself from the road - officially, you should have got off the freeway at the next exit and parked before sorting stuff out.
When something is called CarPlay and actively marketed as a way to use your phone services whilst driving, you would expect it to have a good enough UX that doesnt detract from safe driving to use it.
I mean, sure. Or this could just work how it's marketed to work. Thanks for the helpful reminder!
I have a bunch of "premium" smart outlets(eve energy, some are "THREAD" enabled, all the bells and whistles) all named and connected to things I want to toggle with voice: a small radiator, an air purifier, a humidifier, some accent lamps, a fan, an infrared therapy lamp etc. -Before anyone asks, I ordered the European version of their energy strip for higher wattage stuff-. I give each one a clear and unambiguous name, and still, at least one in ten times Siri will be confused about what I'm asking and turn off EVERY SINGLE OUTLET IN THE ROOM. It's endlessly frustrating. For what it's worth the outlets never de-sync or disconnect like the random amazon ones do so I blame it purely on Siri.
I need better speech to text. I have to repeat my text message 5+ times sometimes when sending it through carplay. I'm not talking about anything long either. A handful of words that my accent just doesn't work with. More than once per day I get fed up enough to pull over and text it.
siri’s speech recognition and intent handling is so comically bad. as of late, for some unknown reason, when i ask siri to favourite the current song while driving it curtly replies that it doesn’t know which speaker i’m referring to. this used to work.
another fun problem is siri not recognising my speech despite me not having a particularly strong accent, speaking slowly, and enunciating. i’ve gotten into the habit putting on a valley girl or bbc news anchor voice while using siri since that usually works.
whoever is in charge of siri needs a reality check, the feature is borderline unusable.
Hey now, Siri’s speech-to-text quality is also worse quality than a vacuum cleaner’s too.
And the vacuum cleaner really sucks
It’s gotta be just barely holding on in some legacy environment right now. I’m surprised they don’t just start over using new tech. Maybe that’s the end goal with on-device inference to sunset cloud Siri.
I think it’s at least partly cost reduction. Push everything to the phone so people have to pay more for phones and they pay less for infra. Win-win.
Otherwise why don’t they let older devices use their server-side private LLM setup?
OpenAI whisper runs on iOS native on device, can't be just costs. https://github.com/openai/whisper/discussions/407
That’s AI though…
I actively don’t want AI, on-device or not. But with the near monopolies on these devices there isn’t a way to vote with your wallet. We’re getting whether we want it or not.
You can have a dumb phone. If you require one for 2fa. Just have one in addition for that purpose. But you can also sms 2fa on many services
I don't want a dumb phone. I want a smartphone without AI crap.
The hard part with 2FA over SMS is that it's no longer considered secure [1]. I want a dumb phone too, but with all the security tools we need (password manager, 2FA apps/tokens, encrypted messaging, etc.) it's becoming less and less an option for me.
I wish there was a middle ground where I could have my phone be dumb enough to keep me from playing on it all the time, but secure enough that it makes sense for me.
[1] https://www.okta.com/blog/2020/10/sms-authentication/ I'm not affiliated with them, just the first article I found on the topic
These stats seem like the reverse of the story in the headline
> A quarter of smartphone owners (25%) don't find AI features helpful,
So does that mean 75% _do_ find AI Feature helpful?
> 45% are reluctant to pay a monthly subscription fee for AI capabilities
Are 55% happy to pay a monthly fee?
>34% have privacy concerns.
66% have no privacy concerns?
This data is in the article
>So does that mean 75% _do_ find AI Feature helpful?
14% find it helpful
>Are 55% happy to pay a monthly fee?
6% are willing to pay
>66% have no privacy concerns?
no stat on this but I think we can assume based on the others that it is not split evenly because that was not the methodology
>>66% have no privacy concerns?
> no stat on this but...
I think we could presume an answer - if the respondents first received a full accounting of how their phones track and record their lives, along with a full list of who is getting that data.
Thanks. That’s clearer!
Ok my next comeback - people are treating AI tools like a musical instrument.
It’s like picking up a guitar for the first time, twanging a few strings and saying ‘Nah, this sounds shit’. Guitars are useless. I’d never pay for a guitar.
Even chatGPT has a learning curve. I save myself hours or days per day using it for all sorts of things. Anyone who says they can’t find a use for AI is just lazy and hasn’t tried hard.
Lazy? Is it a moral failing not to be interested in playing the guitar? I have no interest in guitar music, so why should I "try hard"?
I think you’re missing the point.
There are lots of people who could benefit massively from using AI. Who have no moral objection to it. But they just dismiss it so quickly because it doesn’t instantly, magically make beautiful things for them.
Anyone who cannot get something useful done by AI - who does want to do that - is lazy.
It’s like saying in the 19th century - oh this electricity isn’t really useful for anything much beyond lights. What’s the point?
The poll is very weirdly constructed. It looks like they gave people a set of checkboxes to check rather than asking people to rank their opinions on a scale.
25% checked the box saying they don't find AI tools helpful, but only 14% said they do. Which means 61% checked neither box.
45% are reluctant to pay, but only 6% said they were willing to. So again, 49% simply didn't say one way or the other.
12% straight-up didn't check any box at all.
> 25% checked the box saying they don't find AI tools helpful, but only 14% said they do. Which means 61% checked neither box.
But why include people who didn’t pick anything? Surely they should be ignored and it ends up being 64/36.
Yeah, it's bad. Excluding the people who checked neither is probably better, but it's still not perfect because it leaves out the people who read both options and would have said "undecided" if it were a choice.
It's just a bad poll.
The article provides insufficent data to answer any of those questions, since "don't know" or "undecided" are frequently options in such surveys.
Yet they do provide sufficient information for the headline. AI integrations came in 7th place for considerations when upgrading a smartphone.
Those stats are wrong in another way - everything is about tradeoffs. If the AI is really good I can pay a price to use it - the price depends on how good it is. If the AI is really good I can ignore some privacy concerns - not all but some (indeed in order to work it probably must ignore some privacy concerns)
The graph has multiple options for some of these.
45% unwilling to pay a monthly fee, 6% willing to pay a monthly fee.
You can't say that unless you've seen the test. Was it an Yes/No question, "check relevant to you", or "write how you feel about AI features"? It would change responses drastically
“Turn off the local AI and let me have that extra power” . I think that’s what most people think. I certainly use AI for some tasks, but I don’t need AI everywhere 24/7 in my pocket.
Today in incredibly obvious things...
Smartphones are an absolute graveyard of fads; remember the 3D screen phones, the phones with projectors, and so forth? They generally go nowhere. I suspect 'AI' on phones will be similar.
Overwhelmingly, what people want out of phones is "like my current phone, but with better battery life and maybe a better camera." Previously 'faster' was also a concern, but modern phones are largely Good Enough.
There's a lot we quietly take for granted. Supposedly the NPUs on phones were originally added to let the phone identify objects in your photo library. When shopping for phones, I tend to shop for better battery and camera... but I also wouldn't go back to a photo library without identification.
Yeah, I should possibly qualify; by ‘AI’ on phones I would expect that the average person polled understands LLMs, image generation, that sort of thing. OCR, image classification, etc, I doubt they’re thinking of that when answering this; that wouldn’t even have been marketed as ‘AI’ until a couple years ago.
My favorite one is Amazon's Fire Phone, which had multiple cameras so it could figure out where your face is relative to the screen, which would enable some crazy 3D effect.
Even camera improvements are overblown. My usage and photo quality did not improve much since ~iPhone 6. Taking decent travel photos, selfies, etc - I was happy with the results 10 years ago.
Technically, I understand the difference in the technology, I just don't know who needs that vs who gets excited about new features for a brief moment.
> Even camera improvements are overblown.
I would love to take a photo using my smartphone that doesn't look pixelated, blurry and or over-processed. Maybe asking too much considering smartphone sensors can't compete with DSLRs in some situations, but I'm always baffled with how dark and desaturated some of my photos turn out on my smartphone, as well.
Me too! If anything, iPhone photos became more over-processed over the last decade.
I’m often surprised that my iPhone photo app show old photos in the widget that look stunning and were taken with the iPhone XS and 11 PM. Night photos are better now but day photos stalled a long time ago. I find joy taking some shots with halide or photon in trueRaw where all processing is removed and grain is present. It gives an authentic look … so all HDR stuff is often too much
Few years ago there was usual 'Apple look' on Instagram where every photo was taken at golden hour, all skin blemishes and moles ironed out into oblivion. Everybody was doing it in some form, but Apple was (is?) going furthest, into territory of painting more than representing reality (I compared my Samsung ultra with wife's iphone and its consistently this... but generally all reviews I saw few years ago stated the same).
Now pendulum has swung so all go for 'realistic look', but I expect people actually want rather milder version of above.
Phone photos look OK on phones, but enlarged even the top contender from current dxomark show very much how hardware limits work. Its just not presentable, maybe apart from very bright scenes. Now I wouldn't go bashing phones per se, its marvel what they achieve from those tiny plastic lenses and some CPU time. And they are always there. But any low hanging fruit in phone photography was picked up long time ago by whole market, what lies ahead are slow computational improvements, coupled with very slow increase in size and thickness of camera section of phones to capture more light.
I think some people do genuinely care about the camera improvements; personally anything I’ve used from the iPhone 7 on has been _fine_, but I’m not a demanding camera-user.
These days, manufacturers seem to focus on more niche improvements (like for mobile gamers)
They’ve all been hyped, only to be forgotten as consumers consistently revert to what they actually care about
I would kinda like search that works over my thousands of photos
Google Photos has that. But I guess you mean locally?
Google photos search doesn't work well at all for me.
Even things like "[my dog name] beach" isn't reliable.
Or things like I use photos as notes. It doesn't reliably recall things like cheese when I take pictures of cheese in various stores to remember what is sold where. Not even the name of the cheese; just cheese. Ditto spices.
Don't use the name - "Mr Floppy at the beach" is meaningless, but "dog at a beach" will probably yield a lot more.
I've found google photos search to be pretty good, and if it can't find something usually the map-mode is enough to pin it down (e.g. go to the beach where you took the photo and it shows you the photos from there)
Thanks for the suggestion.
I did just check, and "dog at beach" generates sub 20% recall for me. I go to the beach weekly with my dog, take lots of photos because I'm a dork, and that first query skips many weeks.
Also, I did add my dog as a known / named pet under the explore tab, which is why I thought the name should work.
I can make it work by picking out the beach via geo, but I think the whole thing illustrates how much better this could be. I'd like to be able to get responses to queries like
* [pet name] at [beach X]
* [pet name] with sand on face
* dead seal, or even just dead animal (pics on beach)
* seaglass (recall is poor there too until I manually added to a photo album)
* dent in car
* [spice name] (I take pics of spices to know which stores offer what)
etc etc. The only way I manage the thousands of photos I have now is by carefully sorting into hundreds of albums, which google also doesn't support well.
Amongst the many many deficiencies of the app (which, tbf, does work extremely well as a read-through cache and seems to back things up very well), it likes to surface spotlights of dead people and pets. Which is not at all what I want proactively surfaced.
the most useful ML feature for me is text recognition on photos, helps to copy phone numbers and such
For it to have value they need to effectively apply it for something useful and have it be reliable, not just go hey we have llms. like if i could say siri share my location with my wife and text her "im on my way" oh and send her a pic of what im looking at through my sunglasses. And it worked 99% of the time, and could figure out to do it through google maps because she has android, that would be great.
And, to me, none of that even sounds like AI. Just an assistant app, smart glasses, and google maps not being shit. I can't think of a reason, other than the reasons I already use ChatGPT and the AI object eraser, that I would need AI on my phone. Most of the crap they pump out is just a solution looking for a problem.
Man I really just want email, texting, and a web browser on my phone. More "stuff" to do is an anti feature for me.
Maybe it's just me, but I'd prefer to spend less of my life on my phone in general. Seems like first-party integration of AI features is just a ploy to persuade me to use it more, since the appeal of novel apps or software features has long since died out.
At this point I would pay a premium to have replaceable batteries again.
My Sonim XP3+ has a hard shell replaceable battery and lasts a week and a half on a charge. It's not even a very expensive device, it's under $150! ;) But it is a flip phone running Android Go with no apps...
That said, I think the tradeoffs being made right now are probably the right ones. Apple's latest devices have gone to an electrically released sort of adhesive (versus the older pull-strip removeable adhesive, which is a big step up from the "glue it in" approach many vendors take), and for a given volume, you get more battery if you can rely on the phone to protect it from damage - which is why almost everything with an internal battery uses some variety of pouch cell. They're quite a bit more fragile than the hard-cased batteries, but you get a lot more battery in the volume than you do with the hard cased ones.
As long as it's not incredibly irritating to replace the battery, I'm fine optimizing the daily use thing (battery life in a given phone size) over the once-every-few-years thing (replacing the battery).
Getting someone to replace a battery every 4 or 5 years really isn’t onerous. And for iPhones the magnetic batteries that attach to the back are a pretty effective way to stretch battery life a bunch more hours if you need to.
Field-swappable is a must
The phone manufacturers have to know this, right? they've got market surveys and focus groups and internal dogfooding programs. they all know that a half-baked chatbot experience is going to sell zero phones.
All this AI marketing push has got to be because they think investors are stupid, and they can fool the market into thinking they're doing an AI.
I’m all for AI—if it actually does something useful. But features like swiping to remove unwanted items or ensuring everyone smiles in a photo don’t count. For anything genuinely useful, I’d rather subscribe to an AI service separately than pay a premium to have it locked to my phone. I want access from my PC too.
AI is a means to an end - it's the end that people want (or don't want). For example people like fast and accurate spell check - they don't care whether AI did it or not.
It has been especially frustrating to read about Apple Intelligence on my 2022 iPhone with iOS Safari, where some JavaScript bugs that have been around since 2012 seem to be slightly worse in the latest update, and occasionally the entire app just freezes for no good reason, even if I've closed all other tabs. Not to mention how often I have to reset my entire 2021 iPad because the audio drivers get into a bad state.
I suppose "stability improvements in WebKit" doesn't do much for Apple's stock price compared to SiriGPT. But this is feeling like death-by-1000-cuts: I don't think users are deeply committed to Apple UX/etc. I believe Apple's US market dominance is largely due to burnt fingers around Android's unreliability and annoyances between 2010-2020 (e.g. Google and Samsung not playing nice, badly orchestrated version changes). This was never a permanent state of affairs; Android has stabilized significantly and is more harmonized among competing manufacturers, while retaining its advantages on price and ease of development.
Apple seems complacent on the basics, and is overextending into an AI product few users seem to want.
Do buyers even know what features are there because of “AI”?
For example, on iOS, you can copy text out of photos (https://support.apple.com/en-ph/120004), you can search for photos without entering key words (not that well yet, in my experience, but the results are better than no results), predictive typing apparently uses a language model, modern camera apps do zillions of things to make photos look better, etc.
Neither of those are killer apps, but each does make the device a little bit better.
Everyone sees through a stupid lifestyle theater instantly.
Life isn’t just driving around taking filtered selfies and finding friends faces in albums.
If phones didn’t suck, advanced functions would already be there without “ai”. But they lack any sensible integrations beyond the marketed scenes where clueless people are playing particular job, live and hobby moments.
what's a realistic use case for ai on a mobile phone? i have yet to find myself saying "gee, if only i had ai on my phone, i could do XYZ!"
On iPhone, if I take a picture of a plant or animal, it identifies it for me. It's not 100% by any means, but it's useful enough. I've figured out what were baby plants I wanted vs. weeds. I've figure out species of birds I'd taken photos of with my SLR (ie: phone takes picture of Lightroom editing the image, and is able to identify it from that... I'd prefer there was a way to not require me to take a photo of my monitor, either doing it "live", and/or adding the functionality into the Mac.) For people and pets it can find other images that contain the same subject.
When my daughter was studying Chinese, I could use the live-video translation app and see the lesson text translated to English, and see her hand-written answers also translated to English. I could see this being more broadly useful when travelling, along with live translation of spoken words.
While true your examples are AI, I believe in this case AI is being used in this context to mean LLM-based AI.
I don't know if LLM-based translation is better than previous translation models.
One use case could be improving navigation directions. Right now, map apps provide granular, step-by-step instructions that include unnecessary details, such as how to exit your own neighborhood.
AI could provide more human-oriented direction that focus on key landmarks and decisions rather than every minor turn. For example:
"Hop on 80 West, cross the bridge, take Sir Francis Drake onto 101 South, take the Alexander Avenue exit, don't go through the tunnel, and your destination will be on the right."
Well, I'm still waiting for an AI feature that recognizes my usage patterns, and adapts the system's behavior.
E.g. if it sees that I always reopen an application 2 seconds after the OS kills it in the background, then maybe it shouldn't be killed.
Or if I wake up 3 minutes before the alarm would go off, and take a trip to the toilet, maybe it shouldn't blow up the speaker while I'm frantically pulling up my underpants, but recognize that I'm already awake, or at least wait with the alarm until I'm around the phone again.
Or automatic backlight shouldn't go crazy when I walk in the night under the streetlamps, it should recognize that lamps are coming and going, and that backlight adjustment every 5 seconds is silly and annoying.
I could go on. IMO there is definitely a place for machine learning/AI in phones (and other places too), especially for quality of life thingies. Just nobody is doing them, I guess becacuse these are not as visible as image generation. My credit card has been ready to spend on such developments since at least 2021. One of these days I will have enough of waiting and do it myself, out of spite...
Spellcheck, voice control, voice-to-text, autocomplete and next-word-prediction are all some AI features that are already in use. Voice-to-text could certainly be much better if something like whisper was integrated. I pretty much never actually listen to voicemails, so having a reliable transcription there would be great.
I'd also love to be able to give commands that traverse multiple apps (e.g. take my google sheet and venmo request everyone the specified amount). Most likely this would happen by teaching an AI tool use and having apps expose an API.
I'd love to be able to give voice commands for certain things (e.g. flipping through recipes when my hands are wet) and have the phone be able to do the actual thing I want.
I actually think phones are a much better place for AI since they're so difficult to type on that voice could provide a higher-bandwidth interface.
I'd love it if CarPlay/Siri just could read out stuff it finds on the Internet. Currently, all I can get out of it is "Sorry, I cannot show this to you right now" for basically everything except trying to control multimedia.
At one point, I had ChatGPT working via voice in CarPlay mode (via Shortcuts I think?), but seems like Apple disabled that at one point, for some stupid reason probably.
I used Google lens yesterday to get the artist name of a painting I liked, that was neat.
That's not "on a phone," though. That's just schlepping an image up to the Google data center, and getting a result back. That you're using the phone as an interface to a datacenter doesn't make it "AI on the phone."
The only one I use regularly is object replacement in photos. It's great for editing a street sign out of a picture of the sky or something, especially if you just don't want to dox yourself posting a pic. It's definitely not high quality most times. Just blurry redraw of what the background might look like.
Otherwise, totally with you. No idea why my phone needs AI. I can just open the ChatGPT app if I want to have a discussion with ChatGPT about something. I'm so tired of apps updating to "Add a new AI assistant!" like why do I need to talk to an LLM in most of the apps I use?
This makes sense, AI is a new phenomena and most people aren't using it on their phones as much as they use SM apps, camera. Battery life is one of the major reasons why people upgrade their phone (believe it or not, many don't just replace their battery)
People think of chatGPT now when they hear AI, but for on-device its more like speech recognition, image keyword search, and keyboard next word prediction.
If AI makes some 'thing' look easy, then that 'thing' starts to look cheap.
AI is a technology, not a product feature. I don't care about the underlying technology in my products, I care what they do for me. As long as the feature is great, I don't care if it's made with traditional algorithms, AI, or literal magic--it doesn't matter!
This idea that customers want AI is like saying that customers want applications written in Python. Why would they care what's behind the curtain?
I could not care less about AI on my phone. I don't want to learn to use it nor do I want to give it free reign over my content. If I can disable it on any smartphone I own I will.
Of course, aside from checking email and calendar I don't do a lot of work from my phone.
Zero surprises here. I work with AI daily and am on the same boat.
The only AI thing that excites me on my phone is a battery-friendly way for app developers to run ML models. I don't want an AI assistant that sees everything that i do, but some things like computer vision and translation and OCR are handy to have a super efficient coprocessor for, and I could see interesting apps using ML increasingly. But that's kind of a second order thing. I'm sure apple and them are also trying to figure out how to also let app devs use their strongest foundation models too without every app wanting to download 100GB of weights.
I calculated and found almost no correlation (r=0.2) between battery life and price. Which is quite sad considering it's the most important feature to consumers.
Here's my write-up alongside some interesting dataviz : https://picked.arnaud.at/news/smartphone-data-science
> The biggest motivation for US adult smartphone owners to upgrade their devices is longer battery life (61%)...
Battery replacements are a thing. They're apparently not a common thing, unfortunately, because that might impact new phone sales... and even on devices where it's trivial, people don't seem to do it. A decade ago or so I was doing a lot of Nexus 5 battery swaps for people because one of the battery OEMs was shipping junk and the batteries were shot in a year.
I really wish OEMs would put bigger batteries in phones. It improves everything. Not only do you get longer initial battery life, you can handle far more "battery aging" before things stop working right. You still have a day's battery life (which I expect is what most people actually mean - they want their phone to last the day without thinking about it) even with capacity loss, and a larger battery can have more internal resistance increase (another factor of battery aging) before it sags too badly under the load to keep voltage up.
Based on the fact that easily 90+% of phones I see in the wild have cases on them, physical size and thickness isn't a big factor (and those newer folding screen devices are massive when folded). Another few mm doesn't matter when phones live in purses, men's pockets, or jackets.
... or just go back to a modern flip phone sort of device, get a week and a half battery life, and stop worrying about remembering to charge your phone. ;)
> Based on the fact that easily 90+% of phones I see in the wild have cases on them, physical size and thickness isn't a big factor
Well, people are probably going to put cases on their phone regardless, so the phone has to be thin enough that even with a case it's still a manageable size.
But I agree that battery is where phones ultimately fail most of the time. I have had a Samsung S10 for a bit over 5 years now and it is the battery that is going to force me to retire it and get a new one. Speed, storage, screen quality, camera quality are all perfectly fine for my use case. But the battery rarely lasts a full day with any kind of usage now which is annoying.
I would love a phone with a replaceable battery. But I agree that the average consumer won't bother, they will just get a new one (or they will have upgraded before battery life ever becomes an issue).
So replace the battery? It's not some impossible task.
https://www.ifixit.com/products/galaxy-s10-replacement-batte...
It's slightly annoying to replace, but even if you have a shop do it, it's far cheaper than a new phone.
https://www.ifixit.com/Guide/Samsung+Galaxy+S10+Battery+Repl...
I'm aware it can be done and have looked at the page on iFixit, and may end up doing just that. But as well as looking like a bit of a PITA to do it, I'm not particularly confident my phone will be good as new after I melt the glue and pry it open. My point is that it would be great if replacing the battery was as simple as clicking off (or even unscrewing) the case and taking it out, like it used to be. Granted, I probably couldn't take my phone swimming with me, but I think I could get over that.
People didn't put cases on all phones at the time they could survive small falls and were thick enough to handle.
IMHO the era of "we assume you're going to have a case on this" was cemented in place the instant the camera bump appeared. If you can't set you phone down on a service without it being all tippy thanks to the off center camera bump then you're clearly expected to stuff it into a case that is at least that thick.
Yup. I'd rather the phone be that much thicker, loaded with battery.
I just did a battery replacement on a Pixel 6a. It wasn't terribly expensive, $18 for the whole kit, but the process was very nerve wracking. Unlike Apple's relatively friendly rubber cement Google used some kind of industrial adhesive to glue everything together. You need a good amount of heat to loosen it, but not too much because that will damage the phone. There's a fine line. They do include a helpful strap for removing the battery, but you have that awful glue to deal with. The cherry on top is the slippery little spring clip connector you need to remove without damaging the board underneath.
All in all it made me pine for the days when you could just pop the back off of the phone and yank the battery out to replace it. Personally I think I could live with a 0.8mm thicker phone that still has this feature. Also a headphone jack while you are at it.
> Based on the fact that easily 90+% of phones I see in the wild have cases on them, physical size and thickness isn't a big factor
Everyone says this, but I wonder if the sales data says otherwise, because device manufacturers, especially Apple, REALLY love to advertise how thin their devices are.
Personally, I don't want a super thin device because I feel like the slightest bending force is going to snap it in half. It's absolutely wild to me that people will store their phones in their back pocket.
Apple's mobile division still seems to struggle with the ghost of Jony Ive. At least their laptop division was willing to rethink things with the Apple Silicon transition and go back to things that worked.
The late Intel-era portable Macs were simply broken, because of Ive's apparent obsession with "thin for thin's sake." Or, more probably, "thin for the sake of industrial design awards." They innovated 20% too much on keyboard design with the butterfly keyboard that was simply a bad design. I don't know what else to call a keyboard that reliably breaks after a few years of service. But, bad though that it is, it was even worse because you couldn't replace the keyboard. You had to replace the whole top case and battery, at something like $700, when one key went bad, because the whole thing was integrated to save a mm or two. Not only did they un-solve a solved problem (reliable laptop keyboards), they even un-solved the ability to replace a keyboard when it gets mangled, or something spills in it, or a random failure happens.
But, worse, they went with the USB-C only design (the thinnest port!), meaning that every owner of one of those laptops I knew had a pile of assorted adapters in their bag so they could do things that most people want to do every now and then with a laptop, like drive an external display for a presentation, read a USB-A storage device, or read a micro SD card. Yes, the laptop was thinner, but at the cost of needing to carry extra stuff with you to work around the lack of ports.
And then, because it was so thin, they didn't have space to properly cool the zorching-fast Intel chips they put in them, and the chips would throttle badly while baking the machine.
The M-series laptops, shocking me incredibly, threw all that "design" out the window, and went back to something actually useful. They cool well (admittedly, that's mostly the Apple Silicon barely needing cooling), they fixed the keyboard, and the machines have a useful set of ports on the side for "common tasks with a laptop" (SD, HDMI, headphone, plus the USB-C/Thunderbolt ports - still no A ports, though). And then enough battery to last just about forever.
Anyway, I hope their phone division works this sort of thing out too.
Of course, they are gimmicks to attempt to sell new phones, now that the market has stagnated, many folks even have two or more phones, and hardly need fancy features.
Why can’t they make smaller phones?
They can. It just that no one wants to buy them.
Is it so? Everyone who sees my iphone 13 mini wants it. It’s going at crazy prices in the second hand market
I think the mistake a lot of companies are making with their approach to AI is making AI some "other" interface that you have to interact with explicitly. This chat bot approach is only one paradigm for using AI and it's probably the worst one in most use cases.
Amazon Rufus, for example, falls into this category and it's hard to even remember to engage with it. On the other hand, Amazon is happy to serve me absolute trash search results.
What most people probably want is an AI that can, for example, help organize their calendar transparently. Have a meeting coming up with a client? What if the AI can use tools to go find the latest info about the client? Check their LinkedIn, check their X, check their blog, research their company, give you a tear sheet on topics to discuss based on this in the calendar notes. If a user has to directly interact and instruct the AI, it feels like it defeats the purpose.
This can't be right. I've been hearing about how great AI is for the past year now. It couldn't be that media outlets just report on whatever they think will generate the most revenue.
We'll care more about AI when it's better.
We’re in an era of "subscription fatigue" and smartphone makers adding another monthly fee could be a tough sell
Headsup: CNET is one of those sites playing videos (maybe relevant, maybe not) even on mobile. Be prepared to kill it if you're on metered plan (uBlock Origin can do that on Firefox for Android).
Making things easier to google is IMO the least-impressive use of LLMs. I'm still waiting for a Roomba I can have a conversation with.
I still can't control my iPhone with my voice in the car. Siri doesn't work well for me. Maybe there is another way I have missed, but I'm amazed this functionality is still lacking.
make 5 days battery life - would be real life changing innovations. AI? meeaah? what for? We already have AI for filters like for ages :)
A bit of a chicken and egg problem, isn’t it? Since really useful AI feature’s don’t really exist yet, why would one expect the general public to want it?
Check again in six months.
I would pay extra to -not- have AI in my phone. AI belongs on mainframes, out of my way when I don't want it.
This is an underrated take. Apple, especially, wants to leverage their excellent silicon to do “AI” locally in the name of privacy purely as a marketing advantage, but the tradeoffs there are major (burning my battery instead of data center AC power, plus, my phone will never be better or faster than some big metal GPU with fans), and the privacy issue could just as well be solved by doing a good job designing the server side architecture and educating people that unlike when talking to a person, it’s trivial for a model to “not learn anything from” a conversation, and that that’s in fact automatically how it works unless you’re logging the conversations and purposely feeding them back in.
You can use their cloud instead of local compute if you'd like: https://security.apple.com/blog/private-cloud-compute/
Building a privacy aware cloud is a lot harder than you think it is. Apple is going the whole way with remote attestation, hardware root of trust, software BOM, etc.
https://security.apple.com/blog/pcc-security-research/
The problem is that remote solutions to the privacy problem inevitably involve my trusting someone else's privacy assurances. On-device processing reduces the trust required, although of course it doesn't eliminate it.
The common response to this is to point out that I'm trusting Apple or Google anyway just by using their phones, which is true, but (1) since phones are directly accessible to security researchers, there's more potential to find out about malfeasance if data is exfiltrated from the phone in a way that it shouldn't be, and (2) even if I have complete trust in Apple or Google, I also have to trust that they'll be able to protect my data from malicious actors, and, the less data I give them, the less I have to worry when they fail.
GPT-4o is helpful, and so is NotebookLM. The problem is that "AI" features built into a lot of products aren't that useful.
What surprises me about the story is how weak the effect is, which is stripped by the title mangler: only 25% of users report not finding AI features helpful.
i think the vast majority of all people are meh on AI. i've only ever seen programmers (and grifters of course) get excited about it.
The AI integration I'm looking for isn't to make chatgpt like queries or generate images or find one of my pictures, but to control my damn phone at a much more granular level. Settings are often getting removed or hidden or moved around to the point where controlling your device is a nightmare. Some examples of deeper configuration I'd love to see available:
There's a fundamental tension between what customers want and what providers want.
Providers want to drip feed and charge for 'improvements' indefinitely.
AI is the next drip feed.
The foldable phone thing later in the article is fun, because I actually do want a foldable phone. I'd love to have a smaller phone with a screen that gets bigger for things like watching movies while traveling.
But the thing is, before buying one I borrowed one from a friend and it was absolute shit! The software hardly worked, the hinges felt horrible to open and it was still too big even when folded up. I thought I was going to break it constantly while holding it.
In re: AI on my phone - it's sort of interesting, but really not worth having because it's not smart enough nor capable of being smart enough. It lacks "judgement", which is what I really needs, instead of slavish attention to detail coupled with hallucinations.
I've been trying the iOS Beta 18.2 and here's my impression of Apple Intelligence:
It's not bad, it's really really bad.
The image clean-up feature is utterly useless, and I think it's one of the areas where we can clearly see the difference between a company for whom Generative AI is an afterthought (Apple), and the competition. I paid $1500 for the new iPhone Pro Max and a great part of the deal was the Apple Intelligence support, but frankly, I might as well switch to Android at this point because I'm really disappointed at Apple's take on AI. I'll probably wait until the official Apple Intelligence is introduced but tbh I don't think there will be much improvements over this version.
And as for Siri: It's as stupid as it ever was. I ask it to convert something from lbs to kg and it responds "there's no music playing". If anything, its natural language comprehension has degraded.
Currently, there's no context awareness, so I still can't ask Siri "how do I respond to this email?".
Really, the only thing that "works" is the ChatGPT feature that describes an image you send it. Anything Apple-related is bonkers. It's really embarrassing.
Apple has always sucked in AI and they have never formed the culture necessary to enable and attract top ML talent. They developed Siri years ago and have stagnated since.
As far as the future of compute goes, Meta seems like it has a much more compelling argument with the Orion glasses and their investment in AI.
They didn't even develop Siri, they acquired it. Doesn't seem like they've done all that much with it beyond integrating and marketing it, but I could be wrong.
> Apple has always sucked in AI
Yes, but Apple fans always respond by saying "...but Apple has been using ML in much of the OS for years, you just don't see it..."
I really want Apple to be better at this, but Generative AI is too uncontrollable for a control-freak company.
With all the AI marketing for new phones, I keep yelling to myself "Nobody is asking for this!!" But maybe I'm in a minority group of what I consider "nobody".
It's because current AI is meh on a phone.
Once someone manages to combine Siri's phone integration + ChatGPT Voice mode and it runs locally I will be very interested. Until then, I'm a bit meh on it.
Every time a company shoves AI into their product I see the won't eat cereal meme.
Horse buyers meh on cars, care more about faster horses
Alternate hed: 99% of smartphone buyers have no idea which features are supported by machine inference.
99% of smartphone buyers have no idea what machine inference is.
So why are we surveying them on "AI"?
Because "AI" is used in lots of consumer-facing marketing
So that people can skip reading the article and just post their feelings in the HN comments.